science and the struggle for relevance
DESCRIPTION
Autor: Klass HesselsTRANSCRIPT
Science and the Struggle for relevance
The research reported in this thesis was funded by the Knowledge Network for System Innovations and
Transitions (KSI).
ISBN: 978-90-8891-205-4
Copyright © 2010, L.K. Hessels
Printed by: Proefschriftmaken.nl || Printyourthesis.com
Published by: Uitgeverij BOXPress, Oisterwijk
Cover design by Suzanne Poort, based on a photo collage by Jeroen Kramer (www.kramervanderveer.nl).
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the author.
Science and the Struggle for relevance
Wetenschap en de strijd om relevantie
(met een samenvatting in het Nederlands)
Proefschrift
ter verkrijging van de graad van doctor aan de universiteit utrecht
op gezag van de rector magnificus, prof.dr. J.c. Stoof,
ingevolge het besluit van het college voor promoties
in het openbaar te verdedigen
op vrijdag 19 november 2010 des middags te 2.30 uur
door
laurens Klaas hessels
geboren op 9 september 1980 te groningen
Promotoren:
Prof.dr.ir. h. van lente
Prof.dr.ir. r.e.h.M. Smits
Prof.dr. J. grin
5
contents
chapter 1. introduction 9
1.1 Background, general aim and research questions 9
1.2 Struggles for relevance 11
1.3 Mode 2 knowledge production 14
1.4 Theoretical framework 16
1.5 Methods 21
1.6 outline of this thesis 25
chapter 2. re-thinking new knowledge production: a literature review and a research agenda 29
2.1 introduction 29
2.2 The new production of knowledge: Mode 2 31
2.3 Mode 2 and its alternatives 33
2.4 The reception of Mode 2 42
2.5 The main objections to the Mode 2 notion 46
2.6 towards a research agenda 56
chapter 3. in search of relevance: The changing contract between science and society 63
3.1 introduction 63
3.2 a contract between science and society 65
3.3 a framework to examine the relationship between academic science and society 68
3.4 Struggles for relevance in academic chemistry 75
3.5 conclusion 83
6
chapter 4. Practical applications as a source of credibility:
a comparison of three fields of dutch academic chemistry 87
4.1 introduction 88
4.2 Theoretical framework 89
4.3 Methods: a case study of chemistry in the netherlands 90
4.4 The changing societal contract of dutch academic chemistry, 1975-2005 92
4.5 academic research and practical applications:
a credibility cycle analysis 97
4.6 conclusion: the importance of stakeholders 106
chapter 5. Multidisciplinary collaborations in toxicology
and paleo-ecology: equal means to different ends 113
5.1 introduction 113
5.2 Theoretical framework 115
5.3 research approach and Methods 118
5.4 a changing contract between biology and society 120
5.5 two biological fields under pressure 124
5.6 conclusion 130
chapter 6. Stakeholder interactions in dutch animal sciences 137
6.1 introduction 137
6.2 Theoretical framework 139
6.3 Methods 140
6.4 introduction to the dutch agricultural research system 142
6.5 institutional changes in the research system 143
6.6 relevance in practice: a credibility cycle analysis 148
6.7 discussion 155
7
chapter 7. conclusions and discussion 165
7.1 Brief recapitulation 165
7.2 changes in the struggle for relevance 166
7.3 explaining the similarities among scientific fields 178
7.4 explaining the differences among scientific fields 184
7.5 reflection on central concepts 188
7.6 contributions to the literature about changing science systems 192
7.7 Policy recommendations 198
7.8 conclusion 204
references 207
Summary 223
Samenvatting 231
dankwoord 139
curriculum vitae 241
chapter 1.
9
introduction
1.1 Background, general aim and research questions
The relevance of science is evident yet elusive. today’s major societal issues like ageing, climate change
and depletion of natural resources are so complex that they call for substantial research efforts. But how
to organize these effectively in order to optimize their practical impact? given the intrinsic uncertainty
and unpredictability of academic research activities, both the possibility and desirability of steering its
content are contested. Scholars in science and technology studies discuss the relevance of academic
research in the context of debates about changing science systems. They point to profound changes in
the dominant visions on the societal relevance of science, and also in the interactions of universities with
other organizations expressing these visions. natural scientists themselves are ambivalent about rele-
vance. Promising societal benefits can help them to acquire necessary resources, but it can also restrict
their autonomy and creativity, which count as central values in their work.
The intriguing and important question when and how science can be said to be relevant is the topic
of this thesis. The object of study is the way university researchers deal with potential societal benefits
of their work, or how they struggle for relevance. i will investigate changes in such struggles in eight fields
of natural science. i will analyze the similarities and differences among the dynamics of these different
fields, and reflect on possible explanations for the changes observed.
With this study of the relevance of science i aim to contribute to the understanding of transfor-
mations in the knowledge infrastructure, as discussed in a large and expanding literature. influential
scholars, university managers and policy makers have claimed that the position of universities in modern
societies is changing radically. due to various internal and external factors, academic research is said to
be increasingly connected with practical applications (Ziman 2000, nowotny et al. 2001). it is generally
assumed that the interactions between universities, industry and public organizations have intensified
(etzkowitz & leydesdorff 2000). Moreover, significant changes have been reported in the content and
organization of science, like the growth of ‘transdisciplinarity’ and the broadening of scientific qual-
ity control (funtowicz & ravetz 1993, gibbons et al. 1994). however, as i will show in chapter 2, the
understanding of these dynamics is still limited, due to two problems: first, the empirical evidence sup-
10
chapter 1
porting these claims is not fully convincing, and second, the most influential concept used in this debate
(‘Mode 2 knowledge production’) suffers from conceptual weaknesses that inhibit a proper operational-
ization.
The issue of relevance figures prominently in the literature about changing science systems. one
of the central claims in this literature is that research practices are changing in the sense that research
agendas are increasingly oriented at producing societal benefits, or, in other words, that the relevance
of science is increasingly defined in terms of specific products or policy solutions (see chapter 2 for a
systematic review). This literature points to the relationship between science and its societal context as
an explanation for the increasing societal orientation of research agendas. in this thesis i address these
issues by investigating changing struggles for relevance. to what degree and in what ways do individual
researchers struggle for relevance in their daily work? how has the meaning of relevance changed in
different scientific disciplines? are struggles for relevance facilitated or complicated by the institutional
environment of science? is there a tension between societal relevance and scientific quality? and how
do these struggles vary across scientific fields? By addressing these questions, with empirical data and
theoretical reflection, i aim to make a contribution to the debate about changing science systems.
The main research question of this thesis is:
How to understand the changes in the struggle for relevance of Dutch academic researchers in chemistry,
biology and agricultural science, in the period 1975-2005?
This question can be divided into three sub-questions:
1. What changes can be discerned in their struggles for relevance?
2. how can the similarities among the changes observed in different scientific fields be explained?
3. how can the differences among the changes observed in different scientific fields be explained?
This thesis deals primarily with academic knowledge production, that is, research conducted at univer-
sities and not in public research institutes, industrial laboratories or other knowledge-intensive organi-
zations. The current debate about transformations in the knowledge infrastructure does not deal with
universities only. for example, it is claimed that the interactions between universities, industry and the
government are intensifying (etzkowitz & leydesdorff 2000), and that firms, think-tanks and consultan-
cies play an increasingly important role in the production of knowledge (gibbons et al. 1994). however,
because a shift towards Mode 2 would have important implications for the university and its relation-
ships with its societal environment, i chose the university as an empirical entrance point. indirectly this
approach can yield insights in the changing roles and positions of other organizations in the knowledge
infrastructure as well.
11
introduction
in the remainder of this chapter i will first provide a further specification of the idea of struggles
for relevance (Section 1.2). next i will elaborate on the link between my research and literature about
changing science systems, in particular the notion of Mode 2 knowledge production (1.3). in Section
1.4 i will conceptualize the idea of struggles for relevance in terms of the ‘credibility cycle’, a ‘science-
society contract’ and a ‘research system’. finally i will present my general methodology (1.5) and sketch
the structure of this thesis (1.6).
1.2 Struggles for relevance
The starting assumption of this thesis is that academic researchers struggle for relevance. researchers will
always encounter a certain pressure to position their work in a broader framework. to a lesser or greater
extent, they need to make sure that their work is valuable to society, either directly or indirectly. The
idea of ‘struggles for relevance’ was coined by rip (1988) in an analysis of changing science systems.
he observed that with the emergence of ‘a new layer of institutions, explicitly oriented to “missions”, to
programming, to strategic mobilization’ in the 1970s and 1980s also a shared repertoire for judging rele-
vance had emerged. This development gave rise to struggles for relevance, ‘on top of struggles for funda-
bility’ (rip 1988, p70). in this thesis, i continue with the idea of ‘struggles for relevance’ in a generalized
sense: i consider these struggles for relevance as a universal aspect of academic research with different
manifestations over time and place. Struggling for relevance can involve attuning one’s research agenda
to the needs of societal stakeholders. another form of this struggle is the active transfer of knowledge to
potential users. But, in principle, basic research that does not directly address external knowledge needs
may also be relevant, depending on what counts as ‘relevance’.
indeed the notion of relevance does not have a fixed meaning, but is historically and socially situ-
ated. Building on constructivist approaches in science and technology studies (latour 1987, Yearley
2005), i conceive the meaning of relevance as a product of social interactions among scientists, policy
makers and other stakeholders. in general, relevance refers to the possible (societal) benefits of science.
in the course of this thesis, it will become clear that these benefits can come in many different forms,
ranging from broad cultural values to the development of specific products or the creation of spin-off
companies. after all, ‘society’ is a complex aggregation of numerous individuals and organizations, which
each may benefit from scientific research in different ways. Science can contribute to health care, to
environmental policy making and to industrial innovation, for example. Moreover, even within one sec-
tor the potential benefits of science can vary greatly, and need not be appreciated equally by the various
actors involved. agricultural science can help farmers to raise the productivity of their meat production,
to enhance animal welfare, or to decrease their environmental impact.
12
chapter 1
in more fundamental areas, the term ‘relevance’ can also refer to the significance of a particular con-
tribution in relation to a major scientific problem; this may be called ‘scientific relevance’. in some cases,
scientific and societal relevance can be clearly distinguished as two separate characteristics, but often
there is no clear boundary between them. Many scientists conceive ‘the advancement of science’ as the
most important social or political implication of their research (Small et al. 2008). for example, when
one would ask a biochemist about the societal relevance of his work, he could explain how his research
activities make an important contribution to the understanding of metabolic pathways. The possibility
that this understanding – in turn – could be used by medical researchers to develop new therapeutic
strategies, is then taken for granted. This thesis deals primarily with (struggles for) societal relevance. in-
vestigating the cognitive progress of scientific fields is beyond its central focus. Still, scientific relevance
will receive some attention as well, because it is often linked to societal relevance.
Why would scientists strive for relevance? first, positioning one’s activities in a broader context can
give personal satisfaction1. unravelling the fundamental principles of aquatic ecosystems can be fun
simply to satisfy one’s own curiosity, but may become even more interesting if one knows this enhanced
understanding can help to fight the declining biodiversity. Moreover some scientists experience the suc-
cessful commercializing of scientific inventions as a proof of their personal scientific quality ( Jain et al.
2009). in this way making a contribution to a larger goal could enhance one’s labor satisfaction. Second,
considerations of potential societal benefits can also help to legitimate one’s work to the outside world
(van lente 1993). it is much easier to justify spending large amounts of time and money on scientific ac-
tivities, if they are likely to yield benefits for society. a political scientist interviewed by lamont (2009)
argues that social scientists should pay close attention to policy implications of their work because
‘…it’s the only way that we can justify what we do ... my bias is towards things that are going to make peo-
ple’s lives better ... Because I think intellectuals are leeches if they don’t do that.’ (Lamont 2009 p. 179).
The third possible motivation is strongly related to the second: research of high relevance (in all its
different forms) can provide access to funding and other valuable resources. expected societal benefits
have played a role in research funding since the emergence of modern science (Martin 2003, rip 1997).
Strong alignment of one’s research agenda with the needs of external parties can help to get funding,
either directly or indirectly. in some cases, firms or other organizations directly pay for academic
research addressing their own knowledge needs; in other cases their interest helps to receive support
from innovation-oriented policy instruments (Zomer et al. 2010). Moreover, in some fields stakeholder
interactions provide access to valuable knowledge, datasets or other research materials (Boon & Broek-
gaarden 2010).
1 Personal preference is known to be a very important criterion in scientific problem choice (cooper 2009).
13
introduction
if relevance is so attractive, one may wonder why it would be an object of struggles. There are at least
two reasons why striving for relevance is not a straightforward exercise. first, scientists often struggle
with relevance because relevance is not always for free. researchers may face difficulties when trying to
reconcile the need for relevance with other plans, ambitions and values. Jain et al. have recently shown
that even scientists that engage in commercial activities still cherish their traditional academic identity
( Jain et al. 2009). research issues that are considered highly ‘relevant’ by others do not always appear
to be the most promising in terms of acquiring personal satisfaction, producing high-impact scientific
publications and obtaining peer recognition. This potential tension is particularly important, as it seems
that academic researchers experience a growing pressure for excellent performance in terms of scientific
productivity, resulting from the rise of performance evaluations over the past few decades (Steele et
al. 2006, hicks 2009). earlier studies have shown that it is not uncommon for academic researchers to
work with a double agenda: a public image of their work that selectively highlights the potential contri-
bution to the goals identified by their funding source, and a personal strategy oriented at individual goals
in terms of career advancement or other personal gains (Morris 2000, leišytė et al. 2008). This latter
agenda is often more concerned with fundamental research questions, as these are considered more
challenging (and motivating) and offer better prospects on prestigious publications and on developing a
successful academic career.
Second, struggles for relevance appear when the meaning of relevance is not fixed. Beside a struggle
with relevance, there is also a struggle over relevance. The potential benefits of science are subject to
speculation as they are never certain beforehand. Moreover, due to the plurality of norms, values and
worldviews, there are often disagreements about what should be considered as the benefits of particular
research activities. What one person considers a benefit may be perceived as a threat by someone else.
for example, the question whether government should invest in genetic modification technologies is
highly controversial (Borrás 2006). apart from disagreements about the specific benefits, there is also an
ongoing dispute about the degree to which research activities should be oriented at (short term) societal
objectives (Ziman 2003, lorenz 2008, gibbons 1999, Böhme et al. 1983). in peer review panels evalu-
ating research proposals there are often disputes about the relative importance of ‘social significance’
as a selection criterion (lamont 2009). in negotiations about research contracts or research programs
scientists can be expected to try to convince their stakeholders that their work – even if it has a rather
fundamental nature – is very promising with regard to long term societal objectives. in order to protect
their ‘academic freedom’ researchers may prefer to connect to long term goals, while stakeholders may
want them to address their short term needs (leišyte et al. 2008). eventually, this dispute often boils
down to the question what ‘relevance’ means, or about the mission of academic research.
Based on these considerations i propose the following tentative definition: the struggle for relevance
is the combination of the efforts of scientists to make their work correspond with ruling standards of rel-
evance and their efforts to influence these standards of relevance. depending on the dominant standards
14
chapter 1
of relevance, the possibilities for scientists to optimize the relevance of their work may include attuning
their research agenda with the needs of societal stakeholders and transferring the knowledge to potential
users. But they can also employ rhetorical strategies to present their research in such a way that it can be
argued to fit with dominant standards of relevance (van lente & van til 2008). The second aspect of the
struggle for relevance (influencing standards of relevance) becomes manifest when scientists participate
in public debate or negotiate with specific stakeholders about the benefits that can be expected from
their research. By articulating specific benefits of their work they influence the general view of the value
of their field or discipline, or even of science as a whole. for example, the creation of an ‘innovation-
oriented research Program’ in the netherlands in the area of genomics contributed to the idea that this
field has a high economic potential, and that it is of strategic importance for national competitiveness
(van lente 2006).
as this definition indicates, there are both individual and collective struggles for relevance. individ-
ual scientists try to make their research relevant in various ways. But also research groups and complete
fields and disciplines compete with each other for relevance. for example, spokespersons of catalysis will
publicly argue that their field deserves more support thanks to its contribution to process innovations
in chemical industry. Struggles about the meaning of relevance can also be found on both levels. on a
collective level, leading scientists can influence ruling standards of relevance by participating in public
debate or negotiating about specific research programs. But also individual research proposals and sci-
entific publications influence dominant conceptions of relevance, as they articulate (new) pathways by
which societal benefits can emerge from research activities. in this sense, the efforts individual scientists
make to increase the correspondence of their work with a ruling standard of relevance also affect this
standard. in this thesis both individual and collective struggles are addressed, but the most detailed
analysis is provided of struggles on the individual level, due to the focus on actual research practices.
1.3 Mode 2 knowledge production
after this introduction to the object of study of this thesis, let us now move up one level of abstraction to
the broader scholarly debate it intends to contribute to. This investigation into struggles for relevance fits
in the context of a debate about transformations in the knowledge infrastructure, in particular regar-
ding the possible shift from Mode 1 to Mode 2 knowledge production, which has been claimed in the
literature (nowotny et al. 2001, gibbons et al. 1994). an analysis of struggles for relevance above can
provide empirical insights into the degree to which such a development is taking place in the nether-
lands. Moreover the outcomes of my case studies may also inform the debate with more general insights
into (the factors influencing) the dynamics of scientific fields. The struggle for relevance as conceptuali-
zed above relates in particular to two Mode 2 attributes, reflexivity and quality control. as will be shown
15
introduction
in chapter 2, these are two of the most controversial aspects of Mode 2 knowledge production. neither
of these attributes is supported by convincing empirical evidence. Moreover they seem to be in conflict
with the increasing importance of bibliometric quality indicators, which tend to narrow down the idea
of scientific achievements to strictly academic aspects.
reflexivity refers to the degree to which researchers are sensitive to the broader societal implications
of their work. With their claims about the rise of Mode 2 knowledge production, gibbons et al. argue
that this sensitivity is increasing. Some critics, however, have argued that this trend is probably limited
to a small number of policy-relevant scientific fields (Weingart 1997). Moreover, it seems plausible that
the awareness of (possible) societal impact is present when scientists interact with societal stakehold-
ers in order to acquire funding, but whether this awareness remains active during the complete research
process remains open for empirical investigation. reflexivity is an important aspect of the struggle for
relevance. Making an effort to enhance the societal benefits obviously starts with awareness. So by inves-
tigating the changes in struggles for relevance, this thesis implicitly addresses the issue of (the reported
growth of) reflexivity.
Quality control can be defined as the set of procedures and criteria that constitute the ‘selection
mechanism of problems, methods, people and results’ (gibbons et al. 1994 p. 32). These mechanisms
can be assumed to play a crucial role in struggles for relevance. Practices of quality control, such as
funding instruments and evaluation processes, can be seen as the institutional environment of academic
research, which i conceptualize as part of the academic research system. They provide pressures and
incentives to individual researchers to focus on particular problems, rather than others. The specific cri-
teria guiding these mechanisms determine to what extent it is important to optimize the societal benefit
of one’s research, or to struggle for relevance. gibbons and his co-authors argue that the traditional forms
of academic quality control are increasingly accompanied by novel forms which involve non-scientific
actors and which are based on criteria concerning the societal value of scientific research. however, there
are also indications that academic quality control is increasingly ruled by bibliometric criteria, which
tend to overemphasize the scientific rather than the societal value of research activities (Weingart 2005,
Scott 2007). interactions with possible knowledge users do not seem particularly beneficial for one’s aca-
demic career (goldfarb 2008)(van rijnsoever et al. 2008). What is the actual weight of societal criteria
in relation to other quality criteria, especially in processes other than funding allocation? By investigat-
ing the changing research system and its effects on individual struggles for relevance, i will shed light on
the question whether novel forms of quality control are of increasing importance.
16
chapter 1
Disciplinary differencesone of the main deficiencies in the literature about transformations in the knowledge infrastructure is
the lack of differentiation among scientific disciplines. Some of the most influential writings about this
topic (Ziman 2000, gibbons et al. 1994, Slaughter & leslie 1997) fail to properly distinguish among
disciplines (see chapter 2 for a systematic literature review). it is well-known that science, technology
and innovation studies in general disproportionally draw on evidence about the natural sciences, with a
particular emphasis on a few ‘hotspots’ like high-energy physics, nanotechnology, genomics, biomedi-
cine and other biotechnology fields. This situation carries the danger that observations about a limited
number of (overstudied) fields are generalized to the whole science system, without taking into account
the characteristics of individual fields or disciplines. obviously, observations about natural science can-
not unproblematically be extrapolated to the social sciences and humanities. But also within the natural
sciences there are significant differences in the social organization, cognitive content and types of stake-
holders in society that will influence the struggle for relevance (Bonaccorsi 2008, Whitley 2000, Becher
& trowler 2001). one of the intended contributions of this thesis is to enhance the understanding of
such disciplinary differences and the way they moderate transformations in the knowledge infrastruc-
ture.
1.4 Theoretical framework
This thesis presents investigations into struggles for relevance in dutch university research. to facili-
tate this empirical analysis, an adequate theoretical framework is needed. at first sight, the concepts of
‘Mode 1’ and ‘Mode 2’ may seem attractive, as these concisely summarize several putative trends in con-
temporary science systems. however, these concepts will provide insufficient support for my analysis, as
they rest on a weak theoretical basis (see chapter 2).
My theoretical starting point is a distinction between the agency of scientists and the structures
patterning their behavior. in line with structuration theory i assume that these structures are both the
medium and the outcome of research practices. Structures are the rules and resources drawn upon in the
production and reproduction of social action, but they are at the same time the means for system repro-
duction (giddens 1984). This assumption also resonates with actor-network theory in which structures
are regarded as relational effects that recursively generate and reproduce themselves; succinctly sum-
marized by John law as ‘social structure is not a noun but a verb’ (law 1992).
Building on sociological theories of science, the further theoretical foundation of this thesis con-
sists of two basic assumptions. first, in accordance with the constructivist tradition of science studies
(latour 1987, Yearley 2005, Knorr-cetina 1999), i regard science as a social activity. This implies that
the progress of science should be seen as the product of the socially embedded actions of scientific
17
introduction
researchers, including both interactions with research materials and instruments, and interactions with
other researchers and their wider social context. My second general assumption is that scientists are
involved in a continuous struggle for resources. Building on resource dependency theory (oliver 1991)
and principal-agent theory (van der Meulen 1998, Braun 2003), i assume that scientists depend on their
environment for crucial resources like funding and legitimacy. Scientists compete among each other for
resources, in the form of reputation (Whitley 2000) and credibility (latour & Woolgar 1986).
Based on this theoretical foundation, i will develop and use three central concepts in this thesis
to analyze struggles for relevance: the credibility cycle, the science-society contract, and the research
system.
Credibility cycleto conceptualize the agency of individual scientists in their struggles for relevance, that is, their attempts
to make their work correspond with ruling standards of relevance and to influence these standards, i will
use the credibility cycle (latour & Woolgar 1986). This model explains how struggles for reputation
influence the daily work of individual scientists2. its starting assumption is that a major motivation for
a scientist’s actions is the quest for credibility. Scientists invest time and money expecting to acquire
data that can support arguments. These are written down in articles, which may yield recognition from
colleagues. Based on this, scientists hope to be able to receive new funding, from which they buy new
equipment (or hire staff) which will help to gather data again, etc. conceived in this way, the research
process can be depicted as a repetitive cycle in which conversions take place between money, staff, data,
arguments, articles, recognition, and so on (see figure 1.1). The struggle for relevance can be seen as
part of the credibility cycle. Producing or promising societal benefits may catalyze some credibility
conversions, e.g. the acquisition of funding. however, relevance could also be in conflict with credibi-
lity, for example, when more fundamental research activities lead to more prestigious publications than
application-oriented projects.
The precise form of the credibility cycle can be subject to change. in this thesis i use this cycle as a
heuristic for analyzing academic research practices. This choice implies the assumption that this model
gives a valid and universal representation of the general pattern of scientific agency. during my research,
however, i will also explore whether changes in academic research practices give reason to modify or
expand this model.
Some aspects of the credibility cycle also vary across scientific fields. for example, in fields charac-
terized by a high strategic task uncertainty there is no consensus about intellectual priorities (Whitley
2000). This implies that scientists can earn recognition for a more diverse set of contributions than in
fields with a low strategic task uncertainty.
2 The notion of the credibility cycle will be futher elaborated in chapter 3.
18
chapter 1
Figure 1.1 The credibility cycle, adapted from latour and Woolgar (1986)
The struggles for relevance are primarily located within academic research practices. however, because
‘relevance’ is intrinsically connected to the societal context of science, struggles for relevance are also
shaped by structures extending beyond the immediate research environment. as i have indicated in
Section 1.2, the meaning of ‘relevance’ is the product of interactions of scientists with the wider world.
Moreover, external forces stemming from public policy and knowledge needs of industry (or other so-
cietal stakeholders) influence the degree to which scientists care about practical utility of their work. for
these reasons, changes in the struggle for relevance cannot be understood without taking into account
the dynamics of the organizations in which science is conducted, the institutions governing it, and the
public debate about the value of academic research. in the following chapters i will investigate the chan-
ging structures acting as rules and resources for university research from two different angles, using the
concepts of a science-society contract (chapters 3-5), and a research system (chapter 6), respectively.
although both approaches shed light on the structures influencing the agency of academic researchers,
the first emphasizes discursive structures, while the latter emphasizes institutional structures.
Science-society contractThe first concept used to study social structures influencing the struggle for relevance is the idea of a
‘contract’ between science and society (Martin 2003, elzinga 1997, guston 2000). This concept can
help to grasp the changing societal discourse about academic research. My notion of a contract refers to
the whole of all implicit and explicit agreements between science and societal parties3. it arranges what
science should do (its identity), why it should do this (rationales), and what are the appropriate condi-
tions for science to function well (see figure 1.2).
3 The notion of a contract between science and society will be further elaborated in chapter 3.
Money
Data
Recognition Articles
Staff and equipment
Arguments
19
introduction
Figure 1.2 general composition of the science-society contract
The identity of science, as specified in this contract, refers to its task, which can be defined in its most
general form as the provision of relevant research outcomes. Science’s task is to produce knowledge and to
deliver it in forms like papers, patents, artifacts or educated people. obviously the precise type of know-
ledge science is expected to produce and the form of deliverance vary over time and across disciplines.
The second issue addressed in my conceptualized contract is why science deserves support. Many diffe-
rent (combinations of) arguments are possible, but most of them relate to concepts of societal relevance.
academic research is often regarded as a necessary condition for sustaining a system of higher education.
other prominent rationales for supporting science concern its practical outcomes in terms of com-
mercial products and knowledge of general interest (like expertise which can support decision making
or biological understanding which may lead to medical innovations), either on the short or on the long
term. But basic science can also be seen as a cultural good. The third element of the contract contains
agreements about the conditions under which scientists work and deals with aspects like the scientific
reward system, career opportunities and processes of research agenda setting. although utterances by
scientists, policy makers and other stakeholders form the entrance point for studying this contract, it can
also provide insights into the changing institutional environment of academic research.
in line with the structuration perspective, the ‘conditions’ specified in the science-society contract
act as rules and resources in the conversions of different forms of credibility. These conditions represent
institutions that enable and constrain the agency of individual researchers. The criteria used by research
councils to distribute their funding, for example, contribute to the possibilities of scientists to turn
recognition into money. But shared views about the identity of science or the rationales for supporting
science also act as discursive structures that influence the behavior of individual researchers in their
credibility cycle. for example, public opinion on the importance of disinterestedness may inhibit the
development of close contacts of medical researchers with pharmaceutical firms. in their turn, research
outcomes can also influence or reproduce the contract, for example when they inspire visions on pos-
sible future applications that give rise to new rationales for public investments in research.
Identity
Conditions Rationales
20
chapter 1
Research systemThe other angle from which i will investigate the structural conditions of academic research is by scruti-
nizing the institutions that are part of the ‘research system’ (see figure 1.3). following rip and van der
Meulen (1996), i regard a research system as consisting of ‘research performers (individuals, groups, in-
stitutions), other organizations and institutions, interactions, processes and procedures’ (rip & van der
Meulen 1996 p. 345)4. This system contains universities, related research institutes and funding agencies,
but also governmental organizations, firms and intermediary organizations to the extent that they are
part of the institutional environment. Knowledge users are also part of this system, as they express
expectations about desirable research outcomes, and they sometimes provide research funding. The
selection mechanisms which gibbons et al. (1994) have termed ‘quality control’ can also be regarded as
part of the research system. altogether, this institutional environment provides research organizations
with incentives and constraints to conduct (particular kinds of) research. in this way it gives meaning to
the notion of relevance, and it can either stimulate or inhibit scientists to optimize the societal benefits
of their work.
Similar to the ‘conditions’ of the science-society contract, the institutions of the research system give
shape to certain conversions of credibility. They function as rules and as resources for the ‘agents’ in the
credibility cycle. Simultaneously, funding bodies probably take into account the outcomes of research
practices when formulating their future priorities. in this way, research practices can strengthen these
institutions, but they can also neglect them and put them under pressure. So the research system can
be seen a structure that shapes research practices, but that is at the same time (re)produced by these
practices.
research systems may vary significantly across scientific fields. depending on its cognitive content,
each field has its own stakeholders in society, ranging from environmental ngos to multinational firms
and from farmers to governmental policy makers. These knowledge users provide different pressures
and incentives to academic scientists, exerting stronger or weaker forces on their activities. in some
fields there are traditionally intensive interactions with (potential) knowledge users; in others they are
more distant. another relevant dimension is the degree of strategic task uncertainty (Whitley 2000), or
the relative ‘divergence’ of the search pattern in a given field (Bonaccorsi 2008). in fields with a low task
uncertainty and a convergent search pattern the research system reflects clear intellectual priorities. if
the strategic task uncertainty is high, however, the different actors in the research system disagree about
research priorities and activities diverge in various directions.
4 The notion of a research system will be further elaborated and applied in the analysis of agricultural sciences in chapter 6.
21
introduction
Figure 1.3 The credibility cycle embedded in an academic research system
in this thesis, the main focus will be on the influence of the discursive and institutional structures on
the credibility cycle. My main object of study is the struggle for relevance as performed in the credibility
cycle. i study the dynamics of the research system and the science-society contract primarily in order to
explain the changes in the struggle for relevance. This implies that the influence of changes in the cre-
dibility cycle on the structural conditions as represented by the research system or the science-society
contract is beyond the central focus of this thesis.
1.5 Methods
The general research strategy of my investigation of changing struggles for relevance is a case study ap-
proach (Yin 2003, eisenhardt & graebner 2007). This strategy seems appropriate because my aim is to
contribute to the understanding of transformations in the knowledge infrastructure. This thesis focuses
on three scientific disciplines in the netherlands. My purpose is to generalize my findings to theoretical
propositions about changing science systems, rather than to empirical claims about broader populations.
as has been noted in Section 1.3, one of my intended contributions is enhancing the understanding
of the differences among disciplines regarding the changing relationship with their societal context. a
comparison of various disciplines can be expected to yield fruitful material to this end. i have conducted
three case studies, each dealing with one scientific discipline in the netherlands. My theoretical sam-
pling (eisenhardt & graebner 2007) of disciplines is inspired by Stokes’ typology of scientific research
(Stokes 1997). The two dimensions making up this quadrant model are the relative commitments to
considerations of use and to the quest for fundamental understanding, respectively (see figure 1.4).
Research councils
National academy
Public research institutes
Governmental departments
Firms Association
of universities
University
22
chapter 1
Both seem highly significant in relation to struggles for relevance. for disciplines that are traditionally
strongly inspired by considerations of use it is probably much easier to show the relevance of their work
than for those lacking such a tradition. The degree to which a discipline is concerned with a quest for
fundamental understanding can be expected to correlate with potential barriers for ‘relevant’ research,
for example, when there is a strong tradition of academic publications.
Figure 1.4 Stokes’ quadrant model of scientific research (Stokes 1997 p.73)
Stokes’ dimensions should be seen as continuous rather than dichotomous scales, as research practices
can show commitment to these two goals to varying degrees (Stokes 1997 p. 72). disciplines seem too
diverse to be classified neatly in one of the four boxes. Still the disciplines i have selected for empirical
analysis, chemistry, biology and agricultural science, traditionally vary significantly along Stokes’ two di-
mensions. They were selected based on their variation along these dimensions around 1975, the starting
point of my analyses.
The main share of dutch academic chemistry can be positioned in Pasteur’s Quadrant as it is in-
spired by both the quest for fundamental understanding and by considerations of use. industrial and aca-
demic chemists are known to form dense networks, especially in the netherlands (tijssen & Korevaar
1997, homburg 2003). Still, academic chemistry has a strong academic orientation and is also commit-
ted to producing fundamental knowledge about compounds and reactions. This is illustrated by the rich
tradition of excellent scholarly publications by dutch chemists (Moed & hesselink 1996).
compared to chemistry, most biology was traditionally less directly concerned with practical ap-
plications, especially before the rise of biotechnology. The possible applications of life science research
range across several societal contexts such as agriculture, environmental policy and health care, but these
23
introduction
have traditionally not strongly steered its academic research agenda5. Biology shares with chemistry its
quest for fundamental understanding.
agricultural sciences are (almost by definition) connected to agricultural practice. in the nether-
lands, the Ministry of agriculture has always served as the main funding source for most agricultural
research. Moreover, the dutch agricultural sector is traditionally known as a dense network with strong
interconnections among research, education and extension activities (leeuwis et al. 2006, Maat 2001).
More than the other two disciplines, agricultural research has traditionally been oriented at deliver-
ing practical outcomes. Most fields of agricultural science do not have a strong tradition of scholarly
publications. Their main goal has not been to gain fundamental understanding but to contribute to the
productivity enhancement of dutch agriculture.
for two reasons i have limited myself to natural sciences. first, i can build on a substantial body of
secondary literature. Second, the focus on natural sciences enhances the policy relevance of this study
as these represent the most costly parts of the academic science system. There have been numerous calls
for more studies of the dynamics of social sciences and humanities and some of the issues discussed in
this thesis may create specific problems in these disciplines (Boomkens 2008, van hemert et al. 2009).
however, by limiting myself to the natural sciences i create a body of evidence on a set of disciplines that
share enough characteristics to be fruitfully compared.
in each case, the discipline was studied over the period between 1975 and 2005. i chose 1975 as
the starting point because this marks the beginning of governmental science policy in The netherlands
(Salomon 1980, Blume 1985), which is generally considered as a key-event in the changing relationship
of academic science with its societal context, and because around this time civil society started penetrat-
ing the science system (rip & Boeker 1975, grin 2010).
Within each case i have selected two or three scientific fields (see table 1.1) to represent the breadth
of each discipline in terms of possible societal stakeholders. a part of the structural conditions are shared
among the fields within a particular discipline. Some research councils and evaluation processes are
organized on the level of disciplines. in a sense, however, the various fields can be regarded as separate
case studies as well. in each disciplinary study (chapters 4, 5, and 6) the fields show very diverse dynam-
ics, and comparing the observations on different fields will turn out a rich source of information and
understanding. in the final chapter (chapter 7) this will even lead me to the conclusion that comparing
individual fields is more fruitful than comparing aggregated disciplines.
5 The first research assessment of biology stated that most research could be qualified as purely scientific, not explicitly designed to lead to application (Biologische raad, 1980, Biologisch onderzoek in de subfakulteiten biologie en insti-tuten voor fundamenteel biologisch onderzoek, amsterdam). in the mid-1970s, the main biological research council Bion (Stichting Biologisch onderzoek nederland) only funded ‘purely-scientific’ research which addressed questions of ‘purely fundamental nature, not inspired by any immediate societal issue’ (Bion, 1975, Jaarverslag, den haag, p. 3).
24
chapter 1
Table 1.1 fields selected within each case study and their stakeholders in society
Discipline Fields Main stakeholders
chemistry catalysisBiochemistryenvironmental chemistry
chemical industryBiotech industry / Medicineenvironmental policy
Biology Paleo-ecologytoxicology
oil industryenvironmental policy
agricultural science
animal breeding and genetics (aBg)animal production systems (aPS)cell biology
animal breeding firmsfarmers, agricultural policy(veterinary) medicine
data for the case studies have been drawn from in-depth interviews and documentary analysis. for the
credibility cycle analysis of changing struggles for relevance semi-structured in-depth interviews with
47 academic researchers were carried out6. The respondents’ ranks ranged from Phd-student to full
professor and they were employed at five different universities in the netherlands (see table 1.2). They
were asked questions about their current and past research activities, their personal motivation, and
their experiences and strategies concerning funding acquisition, publishing, scientific reputation, and
performance evaluations. using nvivo (qualitative analysis software), i coded the interview transcripts
in accordance with the different steps of the credibility cycle.
Table 1.2 distribution of 47 respondents over fields, universities and academic ranks
catalysis (9)Paleo-ecology (8)toxicology (7) Biochemistry (6)environmental chemistry (5)animal breeding and genetics (4)animal production systems (4)cell biology (4)
utrecht university (18)Wageningen university (12)university of amsterdam (11)vu university amsterdam (3) radboud university nijmegen (1) eindhoven university of techno-logy (1)leiden university (1)
retired full professor (6)full professor (13)associate professor (10)assistant professor (6)Post-doc researcher (5) Phd-student (7)
My analysis of the changing structural conditions of academic research is based on documents (listed
in the appendices of chapters 4, 5 and 6) in combination with interviews with scholarly experts, and
representatives of firms, professional organizations, research councils and the government. The docu-
ments were collected based on prior knowledge of the authors, tips from interviewees, and the ‘snowball
method’. The selection includes governmental policy documents, reports and strategic plans of research
6 Stefan de Jong and floor van der Wind conducted some of the interviews for the case studies of biology and chemistry, respectively.
25
introduction
councils, foresight studies, evaluations and other important publications about the disciplines addressed.
The findings from these documents were triangulated in interviews with the experts and stakeholders
mentioned above.
1.6 Outline of this thesis
The general aim of this thesis is to contribute to the understanding of transformations in the knowledge
infrastructure by investigating changing struggles for relevance in three scientific disciplines (in the neth-
erlands). The middle part of this thesis (chapters 2-6) was written as a collection of self-standing papers,
which have been submitted to different journals (and one edited volume).
The first step of my itinerary is a systematic review of the current literature on transformations in the
knowledge infrastructure, which is presented in chapter 2. This review served to acquire a general over-
view of the debate about changing science systems, and it was conducted before i had chosen a specific
research question. in fact, the outcomes of this literature study have formed an important input for the
choice of my research topic and formulation of research questions as presented in the current chapter. in
this review the notion of ‘Mode 2 knowledge production’ (gibbons et al. 1994), was chosen as a starting
point, because of its prominence in this debate, both in academia and policy circles. chapter 2 system-
atically reviews the reception of this concept in scholarly literature and compares it to seven alternative
‘diagnoses’ of changing science systems, such as the ‘triple helix’ and ‘Post-normal science’. The review
is concluded with a general research agenda for scholars in Science, technology and innovation Studies,
dealing with the three most controversial features of Mode 2 knowledge production: transdisciplinarity,
reflexivity, and quality control. This general agenda has strongly inspired me in formulating the central
research question of this thesis, which has already been presented above (Section 1.1):
How to understand the changes in the struggle for relevance of Dutch academic researchers in chemistry,
biology and agricultural science, in the period 1975-2005?
as i have argued in Section 1.4, this question deals in particular with the Mode 2 attributes reflexivity
and (novel) quality control.
after this literature review, my next step is a further conceptualization of these issues in terms of my
theoretical framework. chapter 3, ‘in search of relevance’, develops the first two building blocks of this
framework: the science-society contract and the credibility cycle. This chapter also presents a brief case
study to illustrate the usefulness of these concepts. it will explore how changing concepts of ‘relevance’
in the science-society contract influence scientific practices, with a particular focus on the institutions
governing the credibility cycle.
26
chapter 1
The following three chapters use and enrich (elements of) this framework in the analysis of case
studies of dutch chemistry, biology, and agricultural science, respectively. each study addresses similar
questions about the changing struggle for relevance. The specific research questions in each case were
chosen based on their urgency for this particular discipline in relation to existing literature and current
developments.
chapter 4 presents a more elaborate study of the chemistry case. it gives a more detailed analysis
of the changing science-society contract, highlighting in particular the changing funding sources and
the rise of performance evaluations. further it explores the consequences of these developments for
the struggles for relevance of researchers in three different fields of chemistry, guided by the question
whether practical applications have become a source of credibility.
chapter 5 addresses the influence of changes in the contract on the actual behavior of academic
researchers and the content of their work. it focuses in particular on multidisciplinary collaborations
in paleo-ecology and toxicology. Multidisciplinarity is one of the characteristics often discussed in the
debate about changing science systems. it can be seen as a weak type of (and a necessary condition for)
transdisciplinarity, one of the five attributes of Mode 2 knowledge production. This chapter analyzes
the changing structural conditions influencing dutch biologists using the contractual perspective as
presented in chapter 3. next, it investigates how these conditions provide pressures and incentives for
multidisciplinarity in two different biological fields: paleo-ecology and toxicology.
The central question of chapter 6 is: how do changing institutions influence academic research
practices? This chapter deals with dutch agricultural sciences, in particular with three fields of animal
science. These fields have experienced a highly dynamic societal context. dutch agricultural sciences are
under pressure to change the content and the organization of the research in order to contribute to more
competitive and sustainable agricultural production. in this chapter the changes in (institutions of) the
agricultural research system and their consequences for the daily work of dutch animal scientists will be
analyzed.
The thesis is concluded with a general discussion: in chapter 7 i summarize the main findings and
compare them across my three case studies. i relate my empirical findings to the debate about trans-
formations in the knowledge infrastructure, discuss the theoretical contributions of my research and
provide recommendations for science and innovation policy.
chapter 2.
29
re-thinking new knowledge production: a literature review and a research agenda7
Abstract This paper offers a systematic reflection on the gibbons-nowotny notion of ‘Mode 2 knowledge produc-
tion’. We review its reception in scientific literature and compare it with seven alternative diagnoses of
changing science systems. The ‘Mode 2’ diagnosis identifies a number of important trends that require
further empirical effort, but it suffers from severe conceptual problems. it is time to untie its five major
constitutive claims and investigate each separately.
2.1 Introduction
Science systems are said to be in transformation. The last two decades various studies have pointed to a
variety of changes, such as an increasing orientation of science systems towards strategic goals (irvine
& Martin 1984) and the production of relevant knowledge (gibbons et al. 1994, Böhme et al. 1983). a
variety of approaches to understand, explain, and, perhaps, extrapolate such trends have emerged, but
none of them is uncontested. Probably the most famous account of a transformation is the concept of
‘Mode 2’ knowledge production. This notion refers to a set of putative changes that are introduced in
The New Production of Knowledge (gibbons et al. 1994). The book sketches the emergence of a research
system that is highly interactive and ‘socially distributed’. The basic argument is that, while knowledge
production used to be located primarily in scientific institutions and structured by scientific disciplines,
its locations, practices and principles are now much more heterogeneous. Mode 2 knowledge is produ-
7 This chapter has been published as hessels, laurens K., and harro van lente. 2008. re-thinking new knowledge pro-duction: a literature review and a research agenda. Research Policy 37:740-60.
30
chapter 2
ced ‘in the context of application’ by so-called transdisciplinary collaborations. Moreover, scientists are
more reflexive and they operate according to different quality criteria when compared with the tradi-
tional disciplinary mode. The new mode of knowledge production has been coined ‘Mode 2’ and it is
not believed to replace Mode 1, but to supplement it8. table 2.1 gives a summary of the basic claims in a
well-known format.
Table 2.1 attributes of Mode 1 and Mode 2 knowledge production
Mode 1 Mode 2
academic context context of application
disciplinary transdisciplinary
homogeneity heterogeneity
autonomy reflexivity / social accountability
traditional quality control (peer review) novel quality control
in the decade since its launch by Michael gibbons, camille limoges, helga nowotny, Simon Schartz-
man, Peter Scott and Martin trow, the ‘Mode 2’ concept has gained an enormous visibility in the reflec-
tion on contemporary scientific practice. The notion of ‘Mode 2’ is referred to in over 1000 scientific
articles9 and seems to have influenced science, technology and innovation policies10. during the same
period, however, scholars have written numerous critical papers to contest the claims and the use of the
Mode 2 concept, some on a theoretical basis, others supported by empirical data. We think it is time for
reconsideration of the idea of a science system in transformation and we will use the claims and conte-
stations of the Mode 2 concept as an entrance point. to what extent is this concept helpful in describing
and explaining current changes in scientific practice? What does it add to other approaches? What are
the most relevant questions to address when one is interested in the transformation of science systems?
We will follow two routes, one direct, the other indirect. first, the indirect route is to compare and
contrast the Mode 2 diagnosis with a number of alternative accounts of current changes in scientific
practice (section 2.3), such as triple helix (etzkowitz & leydesdorff 2000), post-normal science
(funtowicz & ravetz 1993) and strategic research (rip 2004). We will address both agreements and dif-
ferences between ‘The new Production of Knowledge’ (nPK) - the book in which the notion of Mode
2 has been coined - and the alternatives. This step will make clear about which characteristics of the
science system the different diagnoses make claims and it will show to what extent the claims of nPK
8 ‘This new mode – Mode 2 – is emerging alongside the traditional disciplinary structure of science and technology – Mode 1’ (nPK, p. 14)
9 Scopus search on January 18th 200710 in canada, for instance, the creation of networks of centre’s of excellence aimed at ‘facilitating Mode 2 networks’
(fisher et al. 2001).
31
re-thinking new knowledge production: a literature review and a research agenda
agree with claims made by other authors. The second, direct route is to review and evaluate the numer-
ous reactions to ‘The new Production of Knowledge’. after a discussion of its general reception (section
2.4), a number of critical reactions are addressed (section 2.5). The main objections that we found in the
literature will be grouped under three headings: criticism regarding the empirical validity, the concep-
tual strength, and the political value of nPK. consequently, the strong and weak points of the original
Mode 2 claims can be determined. We will conclude with a statement about the strength and suitability
of the Mode 2 concept and with a list of topics concerning the transformation of science systems that
deserve further study. first, however, we will summarize the two main publications by the creators of the
concept (section 2.2).
2.2 The new production of knowledge: Mode 2
The notion of Mode 2 knowledge production is coined in The New Production of Knowledge (gibbons
et al. 1994). This volume constitutes the outcome of a collaborative research project conducted by six
prominent scholars in the field of science (policy) studies: Michael gibbons, camille limoges, helga
nowotny, Simon Schwartzman, Peter Scott, and Martin trow. The work was originally commissioned by
the Swedish council for research and Planning, frn, aiming to get a view on the future of universities.
The main proposition of the study is the emergence of a knowledge production system that is
‘socially distributed’. While knowledge production used to be located primarily at scientific institutions
(universities, government institutes and industrial research labs) and structured by scientific disciplines,
its new locations, practices and principles are much more heterogeneous. to clarify this assertion the
authors introduce a distinction between Mode 1 knowledge production, which has always existed, and
Mode 2 knowledge production, a new mode that is emerging next to it and is becoming more and more
dominant. five main attributes of Mode 2 summarize how it differs from Mode 1 (see table 1).
first, Mode 2 knowledge is generated in a context of application. of course, Mode 1 knowledge can
also result in practical applications, but these are always separated from the actual knowledge produc-
tion in space and time. This gap requires a so-called knowledge transfer. in Mode 2, such a distinction
does not exist. a second characteristic of Mode 2 is transdisciplinarity, which refers to the mobilization
of a range of theoretical perspectives and practical methodologies to solve problems. transdisciplinarity
goes beyond interdisciplinarity in the sense that the interaction of scientific disciplines is much more
dynamic. once theoretical consensus is attained, it cannot easily be reduced to disciplinary parts. in ad-
dition, research results diffuse (to problem contexts and practitioners) during the process of knowledge
production. Thirdly, Mode 2 knowledge is produced in a diverse variety of organizations, resulting in a
very heterogeneous practice. The range of potential sites for knowledge generation includes not only the
traditional universities, institutes and industrial labs, but also research centres, government agencies,
32
chapter 2
think-tanks, high-tech spin-off companies and consultancies. These sites are linked through networks of
communication and research is conducted in mutual interaction. The fourth attribute is reflexivity. com-
pared to Mode 1, Mode 2 knowledge is rather a dialogic process, and has the capacity to incorporate
multiple views. This relates to researchers becoming more aware of the societal consequences of their
work (‘social accountability’). Sensitivity to the impact of the research is built in from the start. novel
forms of quality control constitute the fifth characteristic of the new production of knowledge. tradition-
al discipline-based peer review systems are supplemented by additional criteria of economic, political,
social or cultural nature. due to the wider set of quality criteria, it becomes more difficult to determine
‘good science’, since this no longer is limited to the judgement of disciplinary peers. however, this does
not imply that Mode 2 research is generally of a lower standard.
in order to emphasize the width of the transformations, the authors of nPK (gibbons et al. 1994)
describe a number of developments in which they are visible such as the commercialization of knowl-
edge, the massification of higher education and the increasing importance of collaboration and global-
ization. The book also includes chapters on the case of the humanities and on the institutional changes
that are involved in the rise of Mode 2 knowledge production.
in 2001, three of the authors of nPK published a second book: Re-thinking Science: Knowledge and the
Public in an Age of Uncertainty (nowotny et al. 2001).11 it can be read as a reaction to some of the criti-
cisms that nPK has received. The authors of ‘re-thinking Science’ elaborate the claims about Mode 2 in
three directions.
firstly, nowotny, Scott and gibbons relate their arguments to sociological literature.They discuss
two accounts of social change that deal with the growth of complexity of society: the Knowledge Society
and the risk Society (p. 10-20). They compare the two along a number of parameters and relate them to
the notion of Mode 2. in addition, the authors relate their work to literature about post-modernism and
co-evolution.
Secondly, ‘re-thinking Science’ extends the argument of Mode 2 beyond the boundaries of the
science system. expanding its meaning, the term Mode 2, here refers to a society consisting of ‘transgres-
sive’ institutions. in a post-modern fashion, the book argues that currently a de-differentiation of the
various societal spheres (state, market, culture) is taking place. These are increasingly fuzzy and blurring
categories that overlap and interact. according to nowotny et al. this development constitutes the back-
ground against which the shift towards Mode 2 knowledge production takes place (p. 29).
Thirdly, the authors make attempts to specify the nature of new scientific practices and discuss ad-
ditional observations of contemporary scientific practice. They describe changes they perceive in various
institutions involved in knowledge production: industrial and governmental research institutes, research
11 in this paper we will use the definitions of Mode 1 and Mode 2 as given in nPK and we will primarily analyse the reactions that this book has received.
33
re-thinking new knowledge production: a literature review and a research agenda
councils and universities. in particular they introduce the concept of ‘contextualized science’ which basi-
cally means that ‘society now “speaks back” to science’ (p. 50). This refers to the demand for innovation,
to new regulatory regimes, and to the multiplication of user-producer interfaces. depending on the de-
gree of importance, one can speak of weak, middle range, or strong contextualization. This development
affects scientific activity not only on the organizational level, but also ‘in its epistemological core’ (p. 94).
The authors claim that Mode 2 (or contextualized) research yields ‘socially robust knowledge’, which has
a different epistemological status than Mode 1 science. Perhaps surprisingly, the participation of a wider
range of non-scientific actors in the knowledge production process enhances its reliability.
2.3 Mode 2 and its alternatives
The Mode 2 diagnosis is popular, visible and contested, but not unique: it appears amongst competing
approaches to study changes in the science system. The first step in our reconsideration is a comparison
with a set of alternatives, which we have identified in a literature study12 (see table 2.2). each gives
an account of current changes in scientific knowledge production and / or the changing relationship
between science and society. We will briefly introduce each of them and discuss the agreements and
disagreements with the Mode 2 diagnosis. Please note that the order of appearance is chronological and
does not reflect any judgement about their relative importance. We will discuss the approaches in terms
of their claims about cognitive changes, organizational changes and science – non science relationships.
The discussion is summarized in table 2.3.
12 The bodies of literature addressed were selected based on their prominence (number of citations) and the degree of apparent similarity with nPK.
34
chapter 2
Table 2.2 alternative diagnoses studied
Concept Aim (descriptive or prescriptive)
Format ‘Main’ publication Number of citationsa
finalization science d / P articles Böhme et al. (1983) 22
Strategic research / strategic science
d (P) diverse irvine and Martin (1984) 58
Post-normal science P articles funtowicz and ravetz (1993)
204
innovation systems d / P diverse edquist (1997) 298
academic capitalism d Book Slaughter and leslie (1997)
315
Post-academic science d Book Ziman (2000) 97
triple helix d articles etzkowitz and leydesdorff (2000)
175
a Scopus search, april 27th 2007
2.3.1 Finalization science
The concept of ‘finalization science’ (Böhme et al. 1983, Böhme et al. 1973) describes and explains the
dynamics of science and its societal function. to some extent, it also contains a prescriptive message.
in the 1970s, the german research group known as the ‘Starnbergers’ (rip 1989), developed a research
programme on science dynamics consisting of case studies of scientific disciplines. The programme has
resulted in a number of journal articles, part of which have been published in german. Most accessible
is an edited volume with contributions of the main proponents of the programme (Böhme et al. 1983).
Based on the case studies, their main claim is that all disciplines follow a general development in which
an explorative phase, a paradigmatic phase, and a post-paradigmatic phase can be distinguished. in the
context of this paper, the last phase is most important. in this phase, ‘finalization’ may occur: theoretical
development that is determined by external factors. When a discipline attains theoretical maturity, it
becomes open to orientation in accordance with external objectives. its further theoretical development
then proceeds along the paths that these goals indicate.
according to the ‘finalists’, more and more disciplines reach this phase. This implies that the relation
between science and society is changing. in this relationship, society is becoming an active rather than a
passive partner, and it increasingly takes a guiding role.
as Weingart has indicated (1997), this observation is comparable to the Mode 2 thesis. four impor-
tant differences need to be mentioned, however. first, the ‘finalization’-approach has a strong empirical
35
re-thinking new knowledge production: a literature review and a research agenda
basis. Second, it clearly differentiates between scientific disciplines: it studies various disciplines sepa-
rately. The claim that the whole science system is undergoing change arises as an inductive conclusion
from the observations of different disciplines. Third, distinct from the emergence of Mode 2 knowledge
production, finalization of disciplines is related primarily to internal rather than external causes. its driv-
ing force is the theoretical maturing development which facilitates social orientation. in contrast, in the
Mode 2 thesis, global developments such as globalization of business and the complexity of policy issues
figure as causal factors. fourth, the Starnbergers are explicitly prescriptive when they speak of ‘normative
finalization’13. They do not merely report the increasing social orientation of science but also provide
policy recommendations. in their writings, one finds a call for ‘social natural science’, science in which
natural norms and social interests are coordinated. given the possibility of social orientation of scientific
research, they argue for setting restrictive conditions that the aims set by scientists must meet (Schäfer
1983).
2.3.2 Strategic research / strategic science
The notions of strategic research and strategic science appear in a variety of sources. They are mainly
used for descriptive purposes, but are often translated into policy goals in a prescriptive way. The term
‘strategic research’ was coined in a policy study (irvine & Martin 1984) and is defined as: ‘basic research
carried out with the expectation that it will produce a broad base of knowledge likely to form the
background to the solution of recognized current or future practical problems’. a striking feature is the
emphasis on basic rather than applied research. This distinguishes this diagnosis from Mode 2, in which
the distinction between basic and applied science has disappeared. Strategic science, however, has inter-
nalized the pressure for relevance while maintaining the (academic) freedom to continuously move to
the most promising line of research. Scientists do not operate in the ‘context of application’, but they do
consider the relevance of their work as a legitimate condition to take into account. There remains a dis-
tance between the actual research and its eventual uptake in the form of solutions to societal problems or
innovations that enhance economic growth.
rip uses the notion of ‘Strategic Science’ to describe an upcoming regime (rip 2004). This regime is
characterized by a recontextualization of science in society. due to the importance of science for innova-
tion and for decision-making, there is more emphasis on strategic research: producing knowledge which
combines relevance14 with scientific excellence. it replaces the regime of ‘Science, the endless frontier’,
in which resources have been available to basic science without requiring clearly articulated promises.
13 This element is particularly visible in the work by Wolf Schäfer (e.g. (Schäfer 1983)).14 in this context, ‘relevance’ refers to application possibilities in either (industrial) innovations or in (governmental)
decision making.
36
chapter 2
rip regards the spread of ‘centres for excellence and relevance’ (rip 2004) and also the commitment of
(entrepreneurial) universities to both regionalism and academic excellence (rip 2002a) as indicators for
the new regime.
2.3.3 Post-normal science
‘Post-normal science’ is a prescriptive approach that is presented in a journal article (funtowicz & ra-
vetz 1993), but it has led to the development of a research community working on the further develop-
ment of the programme. The concept originates from policy relevant science fields and starts from an
acknowledgement of the limitations of rational decision-making. given the complexity of current issues
in (environmental) policy, it argues for a reassessment of the appropriate role of scientific research. in
environmental debates typically ‘facts are uncertain, values in dispute, stakes high, and decisions urgent’
(funtowicz & ravetz 1993). according to the authors, ‘normal science’ in the Kuhnian sense is not an
adequate mode of knowledge production in this situation, as it assumes that problems can be divided
into small-scale problems that can be handled without questioning the broader framework or paradigm.
There is a need for a scientific practice which can cope with uncertainty, with value plurality and with
the decision-stakes of the various stakeholders of the problem at hand. in addition it must have the
capacity to support policy makers taking their time constraints into account. for this purpose the term
‘post-normal science’ has been invented.
The most striking characteristic of post-normal science is public participation. The solutions that
proponents of this model offer generally boil down to engaging stakeholders in decision-making pro-
cesses or in the quality assessment of scientific knowledge production. according to the post-normal sci-
ence view, quality assurance of scientific input to policy processes should be performed by an ‘extended
peer community’ (funtowicz & ravetz 1993). to this end, several frameworks have been developed
that enable dealing with different types of uncertainty, both on the level of model parameters and
assumptions (van der Sluijs et al. 2005) and on the level of societal perspectives (craye & funtowicz
2005) and value diversity (Kloprogge & van der Sluijs 2006).
Post-normal science shares a number of characteristics with Mode 2 knowledge production but plac-
es slightly different accents. common features of both approaches are the increased interaction across
disciplinary and organizational boundaries, additional quality criteria and a greater reflexivity. however,
there is a clear difference in scope. Because post-normal science is only relevant for policy-supporting
research, it does not deal with the university-industry interactions. in post-normal science, corporations
play a role only in as much they are a stakeholder of the policy problem at hand, not because of their
potential role as a knowledge (co)producer. for the same reason, there is no consideration of product
or process innovations, but only for policy innovations or system innovations. Because of its focus on
37
re-thinking new knowledge production: a literature review and a research agenda
the public function of research, post-normal science has stronger similarities with the more recent book
(nowotny et al. 2001) by some of the authors of nPK. it fits the ideal of contextualized research, yield-
ing ‘socially robust’ knowledge.
compared to Mode 2, post-normal science has a more programmatic character. it does not have
a descriptive content in the sense that it reports the emergence of a new mode of research. rather, in a
prescriptive sense, it expresses a need for new modes of knowledge production and aims to contribute to
its fulfilment by developing the required tools.
2.3.4 Innovation systems
Systems thinking in innovation studies emphasizes the importance of interactions and feedback me-
chanisms between all actors involved in innovation, including university researchers, industrial product
developers, intermediary organizations and end-users. The concept of innovation systems is primarily
applied as a heuristic framework, in order to describe and explain the complexity of innovation systems.
in addition, it is used in a prescriptive sense, by arguing for a more systemic innovation policy (Smits &
Kuhlmann 2004). in accordance with the variety of approaches, a diversity of publications is available
on the topic including numerous journal publications and (edited) books. The innovation systems
perspective is applied on various levels of aggregation: national innovation Systems (freeman 1997),
regional innovation Systems (cooke et al. 1997) and technological innovation Systems (carlsson &
Stankiewicz 1991). however, all approaches share a consideration of the interactive nature of successful
innovation processes (edquist 1997).
The innovation systems approaches share with nPK the emphasis on the non-linearity and hetero-
geneity of knowledge production. Both reject the linear model of innovation in which basic research is
translated into applied research, which in turn may result in technological product development (and
subsequent diffusion). in Mode 2, the distinction between basic and applied science does not exist; in
innovation systems such a distinction is conceived to be ineffective. Moreover, the organizational di-
versity of Mode 2 corresponds to the network character of innovation systems. collaborations between
universities and industry, but in particular the role of intermediary research organizations, figure in both
bodies of literature.
a distinctive feature of the systems approach in innovation studies is that it is merely a heuristic
framework rather than a descriptive theory. compared to the Mode 2 thesis, it hardly contains any de-
scriptive claims. While nPK’s authors argue that contemporary knowledge production is heterogeneous
and non-linear by nature, innovation systems literature only argues it should be heterogeneous and non-
linear in order to facilitate fruitful innovation processes.
38
chapter 2
2.3.5 Academic Capitalism
The book ‘academic capitalism’ (Slaughter & leslie 1997) reports the observation of increasing market-
and market-like activities at universities in a set of empirical case-studies15. The authors aim both to des-
cribe and to explain this phenomenon. With ‘academic capitalism’ they refer to two types of activities.
first they point to the increasing (market-like) competition for external funding: grants and contracts,
endowment funds, university-industry partnerships, institutional investment in spin-off companies, or
student tuition and fees. Second they discern increasing market activities: for-profit activity, patenting or
subsequent royalty and licensing agreements, spin-off companies, and university-industry partnerships
having a profit component.
The authors explain this development by two factors. first increasing globalization enhances the
pressure on industry to innovate and causes corporations to turn to universities for assistance. in the
same time, the flow of public moneys to universities is receding. together these factors make universities
more willing to engage in ‘capitalist’ activities. notably, both identified causes are external, in the sense
that they originate outside the science system.
Slaughter and leslie complement their empirical observations with a warning for the risks of the
developments they describe. in all four countries (uS, uK, australia, canada) they have studied,
governments promoted academic capitalism as a means of stimulating economic growth. except for
canada, they all succeeded in developing promoting policies. however, there is no clear indication for
the success of market-activities, as only some universities in the united States manage to make money.
opposite of the potential benefits the authors identify substantial risks for researchers, universities and
their managers. Market(-like) activities can lead to ‘business failure’, to product responsibility, failure to
meet societal expectations (with regard to economic growth and employment) and neglect of students.
for this reason, Slaughter and leslie recommend governments to create incentives for universities to
spend their money in the desired ways in order to avoid a decline in academic education.
‘academic capitalism’ partly confirms the claim of rising importance of Mode 2 knowledge produc-
tion. The market(-like) activities described include (at least to some extent) the attributes ‘context
of application’, ‘organisational diversity’ and ‘novel modes of quality control’. The authors do not pay
particular attention to transdisciplinarity. a curious empirical result is the observation that researchers
are ambivalent with regard to ‘altruism’. They hope that their research will benefit humankind, but this
does not seem to be their first priority (p. 222). from their interviews Slaughter and leslie have got the
impression of researchers being pushed in the direction of academic capitalism, but they do everything
15 There is a large body of literature that deals with the increasing links with industry. however, a lot of studies in this cat-egory start from a firm’s perspective and mainly deal with the potential benefits and costs for industry of collaborating with university researchers (Kaufmann & tödtling 2001, Meeus et al. 2004, Meyer-Krahmer & Schmoch 1998). in the present context, however, we are mainly interested in the consequences of this development for scientific knowledge production.
39
re-thinking new knowledge production: a literature review and a research agenda
they can not to become Mode 2 researchers. They do not show the intention of leaving university as they
prefer to keep the advantages of being ‘state-supported entrepreneurs’ (p. 206).
2.3.6 Post-academic science
in Ziman’s notion of post-academic science, he incorporates elements from several other diagnoses:
Mode 2, academic capitalism and post-normal science. The notion is introduced in a single-author vo-
lume (Ziman 2000), which elaborates on ideas published in his equally successful earlier book (Ziman
1994). Ziman intends to describe and explain a set of developments in scientific knowledge production.
to summarize, post-academic science refers to a ‘radical, irreversible, worldwide transformation in the
way science is organized, managed and performed’ (p. 67). Post-academic science (or ‘post-industrial
science’ science, as Ziman calls it as well) can be characterized by the following five (strongly connected)
elements.
first, science has become a collective activity: researchers share instruments and co-write articles.
Moreover, both the practical and fundamental problems that scientists are concerned with are transdis-
ciplinary in nature, calling for collective effort. Second, the exponential growth of scientific activities has
reached a financial ceiling. The resources available for research seem not to increase much more, creating
a need for accountability and efficiency. Thirdly, but strongly related, there is a greater stress on the util-
ity of knowledge being produced. The success of applying scientific knowledge into products or practical
solutions in some fields has made industry, government and the public impatient with its diffusion rate
in general. There is an increased pressure on scientists to deliver more obvious ‘value for money’. next,
the emergence of science and technology policy has strengthened the competition for resources. in the
resulting situation, competition for real money becomes more important than competition for scientific
credibility. research groups can be conceived as small business enterprises, their staff as ‘technical con-
sultants’. finally, science has become ‘industrialised’: the links between academia and industry become
closer and funding increasingly comes from contract research. This development contravenes the Merto-
nian norms of academic science. due to the industrial orientation a new set of norms can be discerned,
which Ziman labels as ‘Place’: ‘Proprietary, Local, Authoritarian, Commisioned, and Expert’16.
although his approach is primarily descriptive, Ziman is not neutral towards the development of
post-academic science. in a recent paper (Ziman 2003) he draws attention to the ‘non-instrumental roles
of science’, which are threatened in the post-academic era. if science is valued primarily as a mode of
wealth creation, certain functions of knowledge production are overlooked. These include the creation
16 ‘it produces proprietary knowledge that is not necessarily made public. it is focussed on local technical problems rather than on general understanding. industrial researchers act under managerial authority rather than as individual. Their research is commissioned to achieve practical goals, rather than undertaken in the pursuit of knowledge. They are em-ployed as expert problem solvers, rather than for their personal creativity.’ (Ziman, 2000, p. 78-79)
40
chapter 2
of critical scenarios and world pictures, the stimulation of rational attitudes, and the production of en-
lightened practitioners and independent experts. Ziman is convinced that post-academic science is here
to stay; we cannot go back to the old academic model. however, he argues for a fuller consideration of
the non-instrumental roles in the debate about the future of science.
The concept of post-academic science is quite similar to that of Mode 2 knowledge production.
There are no real contradictions between the content of both notions, only some difference in emphasis.
indeed, Ziman refers to Mode 2 in a way that suggests he conceives it as a synonym of ‘post-academic
science’ or at least for the manifestation of that which he calls ‘post-industrial science’ (p. 80). The most
important difference between Mode 2 knowledge production and post-academic science is probably the
scope of the two central notions. Whereas Mode 2 refers to a particular way of conducting and organis-
ing research that constitutes a limited but increasing part of the science system, post-academic science
is a name for the whole science system in its new state. This difference results in a different relation
between the traditional and the new mode of research. While nPK explicitly states that Mode 2 emerges
‘next to’ Mode 1 research and suggest a future in which both develop in co-evolution, Ziman speaks of
post-academic science as a practice that replaces traditional academic research. ‘our exemplar is chang-
ing before our eyes into a new form – post-academic science (…)’ (p. 60).
a similarity between Ziman and gibbons et al. is the loose empirical foundation of their observa-
tions. in both cases, the authors themselves have not gathered any new data. in the same way as in nPK,
Ziman only loosely refers to secondary data, although he does it a little more frequently.
2.3.7 Triple Helix
The triple helix model (etzkowitz & leydesdorff 2000, etzkowitz & leydesdorff 1998, leydesdorff &
Meyer 2006) is based on the assumption that industry, university and government are increasingly inter-
dependent. This implies that these different institutional spheres have to be studied in co-evolution. The
model can be seen as a heuristic forcing researchers to systematically take into account all three spheres
when studying dynamics of knowledge production and innovation. triple helix does not have a uniform
descriptive message like nPK, but it rather constitutes a research program that has yielded a variety of
descriptive claims. its body of literature consists mainly of special issues of scientific journals dedicated
to the triple helix conference series.
The central insight that this approach has yielded is the observation of ‘an overlay of reflexive com-
munications’ between universities, industries, and governmental agencies. according to etzkowitz
and leydesdorff (2000), in most countries there is a tendency towards a knowledge infrastructure in
which these three institutional spheres (academia, state and industry) overlap. in this configuration the
spheres can take each other’s forms and hybrid organizations emerge at the interfaces. The linear model
41
re-thinking new knowledge production: a literature review and a research agenda
of utilization of scientific knowledge is replaced by new organizational mechanisms that integrate market
pull and technology push. Basic research is linked to utilization through series of intermediate processes
such as government initiated programs that facilitate university-industry interaction. The rise of this
configuration is mainly due to the enhanced role of knowledge in our economy and society, and to the
decreasing role of the military.
The role of universities in this configuration is often referred to as its ‘third mission’17. Making a
contribution to economic growth is becoming a central task next to teaching and research18. Within the
triple helix literature, research with this mission is referred to as ‘entrepreneurial science’ (etzkowitz et
al. 2000b, etzkowitz 1998, Kleinman & vallas 2001).
This new role of universities and its new relations with government and industry are roughly in
agreement with the idea of Mode 2 science. especially the context of application and organizational
diversity are apparent. etzkowitz and leydesdorff also confirm transdisciplinarity with their observation
that new disciplines (such as computer science or nanotechnology) arise ‘through synthesis of practical
and theoretical interests’ (etzkowitz & leydesdorff 2000). as has been indicated in the previous section
however they disagree with the view of Mode 1 as the original format of knowledge production. More-
over etzkowitz and leydesdorff prefer to speak of Mode 2 as an ‘emerging’ system emphasising histori-
cal dynamics. in their eyes the current knowledge infrastructure is characterized by mixes of Mode 1 and
Mode 2 (p. 119).
2.3.8 Concluding remarks
table 2.3 summarizes the main findings. The comparison shows that the individual elements of the
Mode 2 diagnosis are not unique. all characteristics that it addresses return in one or more of the other
approaches. nearly all approaches pay attention to the changing research agenda and the increasing
interaction between science and other societal actors. This suggests that these observations are correct,
especially since they are supported by empirical evidence in for instance the ‘academic capitalism’
volume and the ‘triple helix’ corpus. other claims are more distinctive, especially the ones dealing with
methods, epistemology and values. The most striking result of the mutual comparison, however, is that it
shows the exceptionally wide scope of the Mode 2 diagnosis. none of the alternatives deal with as many
characteristics of science as nPK does. it is unclear, however, whether this is a strength or a weakness19.
17 This point is elaborated in literature on the ‘entrepreneurial university’ (etzkowitz 2003, etzkowitz et al. 2000a). although written by one of its main founders, it is not necessarily part of the triple helix corpus.
18 one can argue about the adequacy of the term ‘third’ mission as the authors do not seem to refer to a completely new task but to a reformulation of the second mission.
19 We answer this question in the concluding section
42
chapter 2
Table 2.3 The various diagnoses put emphasis on different characteristics of scientific knowledge production
levels
characteristicsn
PK
Post-normal science
triple helix
Post-academic science
academic c
apitalism
Strategic Science / research
innovation systems
finalization science
cognitive
choice of research agenda (research content) X X X X X X X X
methods (teamwork, transdiscipli-narity) X X
epistemology (socially robust knowledge) Xa X
organi-zational
map of disciplines (transdiscipli-narity) X X
values / labor ethic of scientists (reflexivity) X X
norms of quality control (extended peers) X X X X
external relations
interaction with other societal ‘spheres’ (industry, government) X X X X X X X
incorporation of non-scientific expertise (participation) X
anPK is unclear on this point. The follow-up book ‘re-Thinking Science’ (nowotny et al. 2001), however, does claim that an epistemological transformation is taking place.
2.4 The reception of Mode 2
after the comparison with alternative approaches, we will now focus directly on the strength of the
Mode 2 diagnosis and study its reception in scientific literature. ‘The new Production of Knowledge’
(gibbons et al. 1994) has received over 1000 citations in scientific journals20 and figures 1 and 2 indi-
cate that the number of references per year is still increasing. table 2.4 presents a list of all journals in
20 Scopus search on January 18th 2007
43
re-thinking new knowledge production: a literature review and a research agenda
which 10 or more references were found. as one can expect, journals in the area of science, technology
and innovation studies are dominant, with a light emphasis on journals that are policy-oriented. Some
exceptions in the higher rank deal with management (‘British Journal of Management’), organization
studies (‘organization’), and with policy and planning (‘futures’).
Figure 2.1 number of citations of nPK found in Scopus
Figure 2.2 number of citations of nPK found in Scopus, as a fraction of all articles containing the letter ‘a’, in order to cor-rect for the growth of the number of journals that are included in the Scopus database.
0
20
40
60
80
100
120
140
160
180
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
0
0,00002
0,00004
0,00006
0,00008
0,0001
0,00012
0,00014
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
44
chapter 2
Table 2.4 Scientific journals in which nPK was cited at least 10 times
Journal name Number of references
Science and Public Policy 51
research Policy 40
Scientometrics 38
higher education 36
Minerva 21
research evaluation 20
international Journal of technology Management 19
British Journal of Management 18
Studies in higher education 15
Social Science information 15
technology analysis and Strategic Management 15
Science technology and human values 15
futures 13
Social Studies of Science 12
organization 12
Prometheus 12
r and d Management 12
higher education Policy 10
a closer look at the corpus that is represented in figures 1 and 2 reveals that nPK’s citations can be
roughly divided into two sets. The first set of papers (we estimate about 80%) refers to nPK in the intro-
duction or conclusion section, treating it as an accepted account of the current transformations. in these
cases, the notion of Mode 2 serves to sketch the background for the research that is reported. it helps
either to design a theoretical framework from which research questions are formulated or to discuss the
implications of the findings. for example, Starkey and Madan (2001) use the Mode 2 notion as a theore-
tical framework to discuss current developments in management studies. lee and Bozeman (2005) refer
to the rise of Mode 2 in order to emphasize the importance of research collaborations. lenhard, lücking
and Schwechheimer (2006) cite nPK to sketch the background of their discussion about the future of
45
re-thinking new knowledge production: a literature review and a research agenda
transdisciplinarity21. The content of these articles varies widely, but they have in common that they refer
to nPK in an approving manner, without questioning the validity of its claims.
The second set (roughly 20%), however, does not take the legitimacy of the Mode 2 concept for
granted, but puts its claims to the test. Papers belonging to this category generally dedicate more text
to the issue of Mode 2. They do not cite nPK ‘by accident’, but use it as an essential theoretical starting
point. The following section addresses a number of papers that belong to this second set.
a Scopus citation search22 yielded a list of all scientific articles with a reference to nPK. Based on
source title and article title, a subset of all papers that have been cited 20 times or more was selected for
detailed study. This subset constituted the starting point of the literature study. other papers and book
chapters taken into account were all found by tracing references of this subset. in this way, the most
important23 contributions to the Mode 2 debate should be included (see table 2.5).
21 although the authors do no criticize gibbons et al., their discussion of transdisciplinarity in fact contradicts nPK’s message. Their plea for ‘late integration’ rather than ‘early integration’ amounts to a preference for Mode 1 above Mode 2 knowledge production. The integration of disciplines early in the research project is the central characteristic of nPK’s concept of transdisciplinarity.
22 Search entry: ref(gibbons and “new production of knowledge”)23 in this context, two conditions for ‘importance’ can be distinguished: articles that give extensive reactions to nPK and
that have influenced other scholars (to be measured by the number of citations). note that our focus is on the debate directly linked to the Mode 2 concept. literature about the changes addressed in nPK in fact is much broader, as not all research dealing with these developments necessarily cites nPK. We do not aim to cover all literature dealing with changes in contemporary science systems, but only papers explicitly reacting to the Mode 2 concept.
46
chapter 2
Table 2.5 important reactions to nPK
Authors (year) Title Number of citationsb
godin (1998) Writing performative history: The new new atlantis? 21
Weingart (1997) from “finalization” to “Mode 2”: old wine in new bottles? 41
hicks and Katz (1996)
Where is science going? 35
godin and gingras (2000)
The place of universities in the system of knowledge produc-tion
23
hemlin and rasms-sen (2006)
The shift in academic quality control 0
rip (2002b) Science for the 21st century 3
albert (2003) universities and the market economy: The differential im-pact on knowledge production in sociology and economics
11
Shinn (2002) The triple helix and new production of knowledge: Prepac-kaged thinking on science and technology
19
rip (2000) fashions, lock-ins and the heterogeneity of knowledge production
5
etzkowitz and ley-desdorff (2000)
The dynamics of innovation: from national Systems and ‘’Mode 2’’ to a triple helix of university-industry-govern-ment relations
175
Jansen (2002) Mode 2 knowledge and institutional life: taking gibbons on a walk through a South african university
4
Jacob (2000) ‘Mode 2’ in context: the contract researcher, the university and the knowledge society
10
Pestre (2003) regimes of knowledge production in society: towards a more political and social reading
7
b Scopus search, april 27th 2007
2.5 The main objections to the Mode 2 notion
The criticism of the claims of nPK found in scientific literature is very diverse. We have identified 7
recurring objections, which will be subsequently presented. in our evaluative discussion (5.8 and 6), we
clustered them into three categories: nPK’s descriptive or empirical validity (5.1-5.3), its theoretical and
conceptual strength (5.4 and 5.5), and its political value (5.6 and 5.7). The first type of objection is the
most common, but the other two appear regularly as well.
47
re-thinking new knowledge production: a literature review and a research agenda
in the following, we will discuss the main criticisms to nPK. in some cases we regard it appropriate
to comment on the validity of the objections posed. is the objection indeed a serious problem to nPK?
to our mind, this depends on correct citation of nPK, available empirical evidence and convincing
arguments.
2.5.1 The descriptive validity of the various attributes of Mode 2
table 2.6 lists the authors that comment on the five Mode 2 attributes, indicating the nature of their
comments (positive or negative).
Table 2.6 reactions to the individual attributes of Mode 2.
Mode 2 attributes Godin Weingart Hicks and Katz
Godin and Gingras
Hemlin and Rasmussen
context of application - - +/-
transdisciplinary - - +
heterogeneity - +/- +/-
reflexivity / social accountability -
novel quality control - - +
Context of applicationThe assertion that research is increasingly conducted in the context of application receives relatively little
protest. Weingart (1997) argues that the ‘context of application’ would lack stability, which means that
it will always remain dependent on disciplinary practices. however, this does not contradict the Mode 2
theory as gibbons and his co-authors also expect Mode 1 science to remain present.
according to gibbons et al. Mode 2 knowledge is produced in contexts of application, going beyond
the distinction between basic and applied research. godin (1998) argues, however, that this distinction
(which would disappear in Mode 2) in fact has never existed. fundamental research has always been
inspired by more applied knowledge and applied research has always shown interest in fundamental
understanding of the relevant phenomena. This observation, however, does not imply that basic and ap-
plied research have never been separate domains. godin’s argument does not affect nPK’s claim that the
interactions between basic and applied research are intensifying.
hicks and Katz (1996) test the claim that the locus of knowledge production shifts to the context of
application Their bibliometric analysis, however, does not succeed to either confirm or reject this claim.
in conclusion, the ‘context of application’ remains a complicated concept and there is a lack of clarity
with regard to the difference with ‘applied’ science. a possible solution for this problem can be found in
48
chapter 2
typology of research modes of Stokes (1997). he breaks open the classic dichotomy of basic and applied
research resulting in a quadrant-model, which treats the quest for fundamental understanding and the
considerations of use as distinct variables (see figure 2.3). in this framework, ‘considerations of use’
corresponds to ‘context of application’ and still leaves room for different degrees of being inspired by the
quest for fundamental understanding.
Figure 2.3 Stokes’ model of scientific research (Stokes 1997)
TransdisciplinarityThe claim that science becomes increasingly transdisciplinary receives mixed reactions. The discussion
of this topic is complicated by the lack of universal definitions of inter- and transdisciplinary research.
godin (1998) criticizes the dichotomy between disciplinary research and interdisciplinary research.
according to him, the development of disciplines with specializations and hybrid formations is typical
of any scientific practice. Knowledge production never occurs in isolation; it always involves the employ-
ment of elements from other disciplines. disciplines might acquire some degree of autonomy, but godin
assumes the same could happen to ‘transdisciplinary’ research. This argument would hold for interdis-
ciplinary research. however, transdisciplinarity as proposed by gibbons et al. implies more than only
the cooperation of different disciplines. additional conditions include the co-evolution of a common
guiding framework and the diffusion of results during the research process. The assertion that this type
of knowledge production is currently gaining importance is, in our mind, not sensitive to the criticism
just described.
Weingart (1997) shares godin’s concern that the recombination of disciplines is not a new phenom-
enon. as defined in nPK, transdisciplinarity involves more than that, but according to Weingart, what
the differences are ‘remains vague and ambiguous’ (p. 596). a more serious problem raised by Weingart
is that the difference between the level of program funding and the actual research. research programs
may formulate interdisciplinary or even transdisciplinary problems, but the research carried out under
their headings is often of a disciplinary or multi-disciplinary kind.
hicks and Katz (1996) observe a growth in ‘transdisciplinary’ journals (although disciplinary re-
search still accounts for the bulk of scientific output). however, they employ this term in different sense
49
re-thinking new knowledge production: a literature review and a research agenda
than gibbons et al. according to hicks and Katz, a journal counts as transdisciplinary when it cannot be
classified as belonging to a single field or discipline. They do not take into account at all the epistemolog-
ical and methodological dimensions of the definition of transdisciplinarity as used in nPK. as a result,
the selection made by hicks and Katz includes journals which gibbons and his co-authors would call
‘multidisciplinary’ journals. Their outcomes do prove that interactions between the various disciplines
are increasing, but they can not provide insight into the qualitative nature of these interactions and does
not imply that they are ‘transdisciplinary’ in nPK’s sense.
HeterogeneityMost scholars seem to agree that the heterogeneity of knowledge production is increasing, but they
disagree about the extent.
in a bibiliometrical study, godin and gingras (2000) show that the share of academic publications
that include non-university contributions is increasing, in accordance with the Mode 2 claim of ‘organi-
sational diversity’. in spite of this diversification, however, the presence of universities in scientific papers
is not diminishing at all. in canada the share of papers including a university address has increased from
75% to 82% in the period 1980-1995. from this study it can be concluded that intersectoral collabora-
tion is growing, but universities remain at the centre of knowledge production.
Weingart (1997), on the contrary, argues that the role of think-tanks and consulting firms is negli-
gible in terms of manpower and budgets. This will probably vary for different disciplines; in management
research, for instance, they will be prominent (huff 2000). Weingart’s remark that they remain depen-
dent on academic research does not contradict the nPK claims.
in their bibliometric analysis, hicks and Katz (1996) show that an increasing number of organiza-
tions house authors of journal articles. Between 1983 and 1991, the number of organizations in the
uK participating in scientific publishing has increased in all sectors they studied (hospitals, industry,
non-profit, universities, government and polytechnics), except for research councils24. however, within
the various sectors, hicks and Katz do not discern a uniform trend toward dispersion of the publica-
tion activity. Several sectors (e.g. industry, and hospital) even have become more concentrated. for this
reason, their observation can not be seen as an indicator for the increasing heterogeneity of scientific
knowledge production.
24 This observation may seem to contradict the findings of godin and gingras but does not necessarily imply a difference between canada and the united Kingdom. The co-existence of the findings of the two studies can be explained by as-suming that the share of papers with authors from several sectors has increased, or even more general, that the average number of authors on each paper has increased.
50
chapter 2
Reflexivity / social accountabilityreflexivity and social accountability receives less attention in the criticisms of the Mode 2 concept.
Weingart, for instance, regards social accountability to be mainly applicable to policy-relevant know-
ledge production. in these disciplines he agrees that an institutionalization of reflexive mechanisms is
discernable. Yet, in areas of knowledge lacking an immediate connection to social values and subjective
risk perception (‘high-energy physics, astronomy, and paleontology’ (p.603)), he denies that there is
either a need for or a perceivable rise of reflexivity. This assertion is not supported by empirical evidence,
however. The issue of reflexivity seems to deserve further investigation.
Quality controlThe novel types of quality control probably constitute the most controversial attribute of Mode 2 know-
ledge production.
in agreement with nPK, for instance, Sven hemlin and Søren Barlebo rasmussen (2006) argue
that a shift is taking place from ‘quality control’ to ‘quality monitoring’. Similar to Mode 1, however, the
former remains important. The notion of quality monitoring shares some characteristics with Mode 2.
it is subject to influences of industry and policy, it includes new ‘peers’ (users, consultants, lay persons)
and a greater consideration of ethical and political issues. however, hemlin and rasmussen also add
some elements that are not apparent in nPK: a shift from the assessment of individuals to organizations
and a shift in time perspective from a retrospective judgment of research activities to an ongoing evalua-
tion process during the research progression. The authors also relate the putative shift to organizational
theory and argue for a further study on organizational learning in scientific institutes. however, they do
not provide much empirical evidence for their claims, apart from a couple of illustrative examples.
Weingart and godin are sceptical, however. godin (2000) argues that the scientific criteria are still
the most important. referring to the personal experience of his readers, he states that the success of at-
tracting research funds being dependent on extra-scientific criteria relating to social priorities, relevance
and accountability ‘is still a rhetoric rather than a fact’ (p. 478). in our opinion this assertion does not
do right to the diversity of contemporary funding systems. Probably still a lot of funding allocation is
still ruled by scientific considerations. nevertheless, today a significant share of funding in many western
countries25 depends on societal priorities.
Similar to godin, Weingart (1997) claims that in industrial research marketability and cost effective-
ness have always been present, while in academic quality assessment they are still of minor importance.
even in policy-relevant science that is carried out in the context of application, the scientific standards
assessed by ‘peer review’ remain the most significant measure of quality. however, in contrary to Wein-
25 for instance the european framework Programmes and national innovation policy instruments. examples in the netherlands are ‘technologiestichting StW’, ‘technologische topinstituten’ and ‘innovatiegerichte onderzoekspro-gramma’s’.
51
re-thinking new knowledge production: a literature review and a research agenda
gart’s suggestion on ‘ample freedom for (…) researchers and long-term perspectives’ (p. 603) at corpo-
rations such as iBM, multinationals today seem to be increasingly reluctant to engage in fundamental
research programs. due to globalized competition, industrial corporations have been forced to cut their
budgets and have chosen to cut the costs of long term research activities (de Wit et al. 2007).
in conclusion, the importance of additional quality criteria at universities is contested and remains a
question open for empirical investigation.
2.5.2 Generality of Mode 2 notion
in addition to the detailed comments on the individual Mode 2 attributes, scholars point to limitations
in the empirical validity of nPK’s claims on a more generic level. in this respect, two major problems can
be discerned: the generality of the argument per se and its historical perspective.
gibbons et al. argue that the rise of Mode 2 affects the whole science system: ‘Mode 2 is spreading
across the entire landscape of science and technology’ (nPK, p. 22). ‘These changes appear in the natu-
ral and social sciences but also in the humanities’ (nPK, p. 3).
Weingart (1997) contests these claims. he argues ‘that the features of “Mode 2” (…) are limited to a
fairly small sector of the entire science system’ (p. 608). he claims that some Mode 2 attributes (context
of application, transdisciplinarity) make sense only for science which is close to policy-making such as
environmental research. The subset of academic research from which he believes gibbons et al. draw
their evidence, represents only a fraction of the entire science and technology system. features like ‘un-
certainty of knowledge, complexity of subject matter, policy orientation and value-ladenness’ (p. 600),
which characterize the sector of technology assessment, risk research, and environmental and climate
research, cannot be generalized. Therefore, Weingart sees no reason to believe Mode 2 will extend to all
other areas of science.
godin (2000) rejects the generality of nPK’s approach, too. in particular, he argues that ‘the social
sciences, as well as the humanities, have always been of Mode 2, much more than has been the case for
the natural and physical sciences’ (p. 472).
albert (2003), however, has shown that there is no observable trend towards Mode 2 in the sociol-
ogy and economics departments of two canadian universities. Based on interviews with scientists and
a study of their publications, he states that there is no tendency towards problem-oriented research, but
rather a predominance of Mode 1 knowledge production. albert concludes that his findings demon-
strate that academic research can not be seen as a homologous unit. he argues that one should take into
account the heterogeneity of scientific disciplines and of the various research ‘regimes’.
critics also mention the neglect of national contexts. Shinn (2002) claims nPK fails to ‘recognize
that the university, business and government all function in a national setting’ (p. 610). he argues that
52
chapter 2
scientific disciplines and specialties operate differently in different national institutions. in spite of
current globalization, the national component of the organization and work of science is still apparent.
referring to the literature on national systems of innovation, Shinn states that, even in europe, national
science policies are still of great importance.
The claim of nPK of a general move to Mode 2, therefore, is denied by several authors on different
and sometimes contradicting grounds. This implies in any case that the specificity of disciplinary devel-
opments needs be taken into account much stronger.
2.5.3 The long-term historical perspective
nPK’s second generic limitation indicated in literature concerns its historical perspective. referring
to historical studies of sciences, several scholars (etzkowitz & leydesdorff 2000, rip 2000) claim that
at least some of the attributes of Mode 2 knowledge production have always been present in modern
science.
rip (2002b) rejects the view of Mode 1 as the original type of research. We should rather see it as
historically located: it emerged during the course of the19th century and became locked in during the
1950s and 1960s. features of Mode 2 such as heterogeneity and transdisciplinarity are not new; they
were already present in the ‘renaissance melting pot’ (rip 2000) before the birth of modern science.
etzkowitz and leydesdorff (2000) also contest the newness of Mode 2 and use similar arguments.
referring to historical studies of science they claim that it is not Mode 1 but Mode 2 which is the origi-
nal format of science, as, in the 17th century, research focused on practical problems. ‘Mode 2 represents
the material base of science, how it actually operates. Mode 1 is a construct, built upon that base in
order to justify autonomy for science, especially in an earlier era when it was still a fragile institution and
needed all the help it could get’ (p. 116).
Similarly, Pestre (2003) argues that elements of Mode 2 have always existed in modern science.
Knowledge producers have never isolated themselves in an ivory tower, but have always paid attention
to the interests of states and economic elites relating to science. Moreover, ‘science has always directly
contributed to, and has been a major resource for, changes in social ideologies’ (p. 250).
2.5.4 The coherence of the concept
The Mode 2-authors describe Mode 2 as a stable entity, with a specified set of characteristics. This
implies that the notion of Mode 2 is coherent in the sense that the various attributes mutually correlate.
‘These attributes, while not present in every instance of Mode 2, do when they appear together have a
53
re-thinking new knowledge production: a literature review and a research agenda
coherence which gives recognisable cognitive and organisational stability to the mode of knowledge
production’, gibbons et al. declare (nPK, p. 8). critics seriously question this assumption, and the criti-
cisms discussed above illustrate this. if the evidence for the diverse attributes of Mode 2 varies, and some
receive more assent than others, one can wonder about their mutual relations. Possibly the claim of the
nPK authors about the rise of Mode 2 should be divided into five different claims about five distinct
trends in contemporary science.
according to rip (2002b), the separate features that gibbons et al. describe ‘are clearly visible, but
one might question their overall thesis that these add up to a new mode of knowledge production’ (p.
104-105). he doubts whether the features together have enough stability to make it appropriate to speak
of a new research mode.
godin (1998) confirms two of those claims, but he rejects the other three. he agrees that the hetero-
geneity of the science system has increased and that researchers are more socially accountable. But in his
view both the ‘context of application’ and ‘transdisciplinarity’ have always existed, while novel criteria of
quality control are not yet apparent. Similarly, Weingart (1997) considers all attributes to be present in
policy-relevant science, except for novel quality criteria.
2.5.5 Theoretical underpinning
Shinn (2002) discusses a second conceptual problem. he is concerned about the lack of theoretical
underpinning of nPK’s sociological framework. in Shinn’s reading, ‘anti-differentiationism’ is the central
feature of nPK’s approach, as it blurs the boundaries between academic, technical, industrial, political
and sociological institutions. however, this central idea ‘is never buttressed with sociological theory,
concepts or models’ but it ‘stands as a free-floating, unintegrated component’ (p. 604). nPK does not
account for ‘how differentiations have operated in the past, how and why they would have eroded, and
what their putative demise implies for sociological theory’ (p. 611).
This criticism is not completely fair. The fact that nPK does not talk about the past indeed is a severe
limitation that calls for further study. however, although its treatment of the mechanisms may not be
sociologically sound, it does provide an account of the why and how of the erosion of differentiations.
it refers to causal forces in the supply and in the need of knowledge and it offers an explanation of the
putatively increasing interactions across disciplinary and institutional boundaries. explicit relations with
sociological theory are indeed absent in nPK, but these receive some attention in the follow-up book
re-thinking Science.
54
chapter 2
2.5.6 Mode 2: wishful thinking?
Several authors complain about the uncritical blend of descriptive and normative content in nPK.
according to godin (1998), the talk of Mode 2 is more a political ideology than a descriptive theory.
he associates the message with a ‘polarized rhetoric’, which ‘denounces many of the characteristics of
contemporary research and training in the name of social and political desiderata which are themselves
in exact opposition to characteristics of traditional academic research’ (p. 479). The lack of empirical
foundations makes this a dangerous situation from a policy standpoint as it can easily raise the impres-
sion that the current research system needs to be replaced. godin warns that some readers of nPK may
conclude that the old system and the old academics are wrong and that a new type of research would be
better than traditional academic research.
The comments made by Shinn (2002) are similar: ‘instead of theory or data, the new Production
of Knowledge – both book and concept – seems tinged with political commitment’ (p. 604). Shinn
perceives the authors as being in favour of a new cognitive and social order. in his reading, they aim to
support Mode 2 by persuading others to believe in its importance and desirability.
This relates to what Weingart (1997) calls ‘performative discourse’. he also has the impression that
the idea of Mode 2 is part of a normative program rather than an empirical analysis (p. 608), but in this
remark he mistakenly fails to distinguish between Mode 2 and post-normal science. in fact, the latter
indeed contains strong normative elements, but – as we have argued in Section 2.4 – this is exactly the
main difference from the former.
Some commentators have indeed treated nPK as a prescriptive rather than a descriptive theory. for
instance, the criticism uttered by Jansen (2002) and Jacob (2000) is more concerned with the problems
of bringing Mode 2 into practice than with the question whether the change is actually taking place. in
contrast to the previous critics, however, these do not contest the fact that nPK may bear a normative
dimension, but question the validity of the normative content itself.
2.5.7 Lack of future outlook
a minor comment to be mentioned here is expressed by Weingart (1997), who accuses gibbons et
al. of not being clear with regard to the persistence of Mode 1 knowledge production (p. 593). nPK
indeed is somewhat ambivalent in its future outlook. on the one hand, the authors acknowledge that the
disciplinary forms of cognitive and social organization continue to be prerequisite of identity, as happens
during education and training. on the other hand, they expect that eventually ‘Mode 1 will become
incorporated within the larger system which we have called Mode 2’ (p. 154), but this statement does
55
re-thinking new knowledge production: a literature review and a research agenda
not receive any explication. What gibbons et al. exactly expect to remain of Mode 1 research, therefore,
remains vague.
2.5.8 Concluding remarks
clearly, the empirical validity of the Mode 2 claims is limited. on a generic level, critics convincingly
indicate two major problems. first, the nPK authors disregard the diversity of science and second, their
historical view is mistaken. Moreover, some of the Mode 2 attributes are heavily disputed. in particular
empirical evidence to show the rise of reflexivity, transdisciplinarity, and new modes of quality control is
lacking.
concerning nPK’s conceptual strength, one can argue that more links to sociological theory are
required. We consider the problem of a lack of coherence more serious. given the fact that the different
attributes of Mode 2 receive assent to varying degrees, there seems no compelling reason to tie them
together under a common heading. Whether Mode 2 has ‘recognisable cognitive and organisational
stability’ is highly questionable.
in our opinion, the comments regarding the normative message of nPK do some injustice the
content of the book26. a careful reading reveals that the authors do not have explicit normative inten-
tions. on the first page, gibbons et al. already mention that their intentions are descriptive rather than
normative: ‘no judgement is made as to the value of these trends – that is, whether they are good and
to be encouraged, or bad and resisted…’. The remainder of this sentence, however, can explain some of
the confusion: ‘…- but it does appear that they occur most frequently in those areas which currently
define the frontier and among those who are regarded as leaders in their various fields.’ Strictly speaking,
this sentence does not qualify Mode 2 as better than Mode 1, but it does hint in this direction. So there
is some rhetoric, and as rhetorics about science can lead to social reality (van lente & rip 1998), the
critics have right to complain. This means, thus, that the nPK authors could have been a bit more careful
in distancing themselves from normative claims, that is, if that was their intention.27 in addition, more
clarity with regard to the future of Mode 1 knowledge production is desirable.
26 neither godin, Shinn nor Weingart support their objections with citations from the book.27 note, however, that even when the authors did not have any normative ambitions, the notion of Mode 2 could be taken
up by others as a direction to follow. texts have a life on their own, beyond the reach of authors, and may lead to self propelling dynamics.
56
chapter 2
2.6 Towards a research agenda
our comparison with alternative diagnoses of the dynamics of contemporary science systems shows
the particularly wide scope of the concept of Mode 2 knowledge production. The diagnosis expressed in
nPK (gibbons et al. 1994) includes statements about changes on the cognitive, organizational and the
societal level. Whatever one may think of the adequacy of its analysis, this must be regarded as an accom-
plishment. Thanks to its breadth, nPK has created a forum to discuss a wide range of putative trends.
There are certainly big differences in aims and scope between the various diagnoses that we have
addressed in Section 2.4. an important difference is their balance between descriptive and normative
content. While the ‘post-normal’ science literature shows a clear normative orientation, others primar-
ily limit themselves to reporting observations. another difference is between heterogeneous research
programmes (such as triple helix and innovation systems) and well-defined analyses (such as nPK and
academic capitalism).
however, as displayed in table 2.3, the content of the various accounts show strong similarities.
The claim that the content of scientific research agenda is currently changing recurs in all diagnoses: all
address a turn towards more relevant research, research that (sooner or later) may lead to applications
in the form of innovations or policy. furthermore, all approaches point to more interactive relationships
between science, industry and government. in conclusion, at least two characteristics of knowledge
production that nPK claims to be undergoing change find strong resonance in the alternative diagnoses.
in other words, there is some degree of consensus on these points.
The other characteristics that are addressed in the Mode 2 diagnosis are much more contested. The
conviction that the shift towards more relevant knowledge production also involves a change in the
research methods is hardly shared by other diagnoses. The same holds for the claims that the disciplinary
map is undergoing profound changes and that there is a rise of novel modes of quality control. to con-
clude from this comparison then, that these three claims are wrong, would be too quick. it does point,
however, to the need for empirical evidence for the putative changes.
The Mode 2 claims have received mixed reactions: hundreds of papers cite nPK affirmatively and policy
makers use the arguments, but there is also serious criticism. an analysis of the receptions of nPK in
the literature has yielded a list of seven objections, which we divided into three categories: the empirical
validity, the conceptual strength, and the political value of nPK.
Empirical validity1 There is a lack of empirical evidence for the rising importance of the attributes of Mode 2 (Weingart
1997, godin 1998, hicks & Katz 1996).
57
re-thinking new knowledge production: a literature review and a research agenda
2 The long-term historical perspective is incorrect: the view of Mode 1 as the original type of know-
ledge production is contested (etzkowitz & leydesdorff 2000, rip 2000, Pestre 2003).
3 The universality of the claims is not justified: in contrast with the generality of nPK, scholars expect
the dynamics to be different in different national contexts and in different scientific disciplines
(Shinn 2002, tuunainen 2005, albert 2003).
Conceptual strength4 The necessary coherence of the concept is questionable: there might be a lot of multidisciplinary,
application oriented research that does not show organizational diversity or novel types of quality
control (rip 2002b).
5 The claims lack a theoretical underpinning and references to sociological theory (Shinn 2002).
Political value6 The authors seem to implicitly support the observed trends (Weingart 1997, godin 1998, Shinn
2002).
7 The book lacks a proper future outlook (Weingart 1997).
These seven objections point to problems with the content as well as to problems with the form of nPK.
The last three accusations, for instance, merely deal with the form that nPK’s authors have chosen to
present their message. in fact, they indicate that nPK does not always meet the standards that a scientific
reader expects. although many critics have treated it as a scientific theory, we rather suggest conceiving
it as a manifesto28, in which the authors are more concerned with getting the message across than with
building sociological theories or with carefully distinguishing their observations from their opinions.
however, this observation does not necessarily affect the validity of the descriptive and conceptual
content of nPK. even the most carelessly written manifesto can still point in the right direction. and,
in the case of nPK, even if both complaints about its political value were correct, the concept of Mode
2 can still be useful: they affect the appreciation of nPK as a book, but not necessarily as a descriptive
project. The same holds for objection 5. Yes, the lack of theoretical underpinning definitely constitutes a
limitation of nPK. nevertheless, it does not automatically have implications for the accuracy of its diag-
nosis. The fact that the Mode 2 concept was introduced without references to sociological theory does
not imply that it cannot be related to theory. for the concept to be viable, however, it would be required
to make these references later to avoid it becoming a theoretical ‘island’.
The first 4 objections are more severe threats, especially the noted lack of coherence. Probably least
threatening is the mistaken historical perspective, as this weakness can be corrected. in this respect,
nPK is simply wrong, as it gives a too linear account of the historical dynamics of scientific practice. The
28 note that the volume was originally written for an audience of policy makers rather than scientists.
58
chapter 2
suggestion that a traditional disciplinary mode of research is gradually giving way to a more interactive
mode is not historically correct. for example, intimate interactions between science, invention and
entrepreneurship were already important in the British industrial revolution (freeman 1997). another
example is the well-known steep increase in prestige of and available funding for basic research in
disciplines such as chemistry and physics just after WWii provides another illustration. inspired by the
presidential advice by vannevar Bush (1945), western economies devoted large amounts of money to
basic research, as conducted in universities29. This points to a strong increase - or even stronger: a ‘lock-
in’ (rip 2000) - of Mode 1 at the expense of research with Mode 2 characteristics. We should not regard
the changing modes of research as a one-dimensional development. Yes, it is conceivable that currently
Mode 2 knowledge production is gaining importance in comparison with Mode 1. But to view of Mode
1 as the ‘traditional’ mode, and Mode 2 as the mode which introduces radically new characteristics to
scientific practice, is incorrect. as has been suggested by Martin (Martin 2003), it may be more ap-
propriate to speak of ‘shifts in the balance of Mode 1 and Mode 2 over time’. research is needed, then, to
specify the historical contingencies in the Mode 2 concept.
Problem 1 indicates a serious weakness of nPK: its lack of empirical evidence. We concluded that
some claims are readily shared amongst alternative approaches. of the five attributes of Mode 2, the
‘context of application’ and ‘heterogeneity’ receive assent both in the alternative diagnoses and in the
direct reactions to nPK in scientific literature. for the other three (transdisciplinarity, reflexivity, and
novel modes of quality control), however, neither confirming nor falsifying evidence is available. further
empirical research is required to decide whether nPK’s claims about these points are appropriate.
The idea of transdisciplinarity is contested. in their reactions to nPK, several authors have addressed
the issue of interdisciplinarity with theoretical comments (godin 1998) or empirical investigation
(hicks & Katz 1996). however, the question of nPK’s concept of transdisciplinarity is still open, be-
cause this includes additional features when compared to the concept of interdisciplinary research. is the
integration of disciplinary research elements as dynamic as gibbons et al. argue? numerous publications
are available on the issue of transdisciplinarity (lenhard et al. 2006, Pohl 2005, despres et al. 2004),
but these do not demonstrate the actual diffusion of this mode of research30. a question to address in
an empirical study is how often the research outcomes are indeed communicated during the process of
knowledge production.
29 ironically, the original report did emphasize the importance of interconnections between basic research and other parts of innovation processes, too. however, the report has generally been interpreted as a propaganda document advertising the importance of funding basic research and viewing applied research as ‘second rate’ (Shapley & roy 1985).
30 it must be noted that various definitions of transdisciplinarity are available, some in disagreement with nPK (lawrence & despres 2004). for some (regeer & Bunders 2003), the distinctive characteristic of transdisciplinary research is the inclusion of knowledge of non-scientists. Strictly speaking, however, this feature is more related to what gibbons et al. call ‘heterogeneity’ than to their concept of transdisciplinarity.
59
re-thinking new knowledge production: a literature review and a research agenda
The claim of nPK that reflexivity and social accountability tend to increase also merits more
research. Slaughter and leslie (1997), for instance, report that the ambition to enhance human welfare
does not seem to be a first priority for scientific researchers. it remains to be seen to what extent the (po-
tential) relevance of their work influences the choices researchers make about problem choice, research
design and methods. This raises questions about the reflexivity reported by gibbons et al. can one
discern an increased awareness of possible societal effects on the actual laboratory floor or is it only vis-
ible during interactions of researchers with their societal ‘stakeholders’? following Weingart’s criticism
(Weingart 1997), one can wonder whether the trend of increased reflexivity is limited to policy-relevant
sciences. There are many other fields that appear to have a stronger orientation towards commercial ap-
plications, but it is uncertain whether this makes the scientists involved more reflexive.
about the possible new modes of quality control the evidence is mixed. Some studies confirm the
change (hemlin & rasmussen 2006) but others (Weingart 1997, godin 1998) reject it. in Swedish
funding agencies for technical research, applicability has become an important criterion (Benner &
Sandstrom 2000), but the impact of this change is still uncertain31. after a comparison of funding prac-
tices in Sweden, the uK, norway, canada and the uSa, Benner and Sandstrom conclude that, although
new criteria such as utility and demands from ‘customers’ have been added, research councils preserve
their core orientation: the collegial control and evaluation of research (Benner & Sandstrom 2000). in
the netherlands, societal relevance is a structural element of the evaluation procedure of research group
performances32. however, it is unclear how important this specific criterion is and how it should be
measured. recently a number assessment tools have been developed and tested which address criteria
for societal relevance. examples are the ‘societal quality research profile’ (SQrP) in health research
(Spaapen 1995) and the ‘research embedment and performance profile’ (rePP) of agricultural sciences
(Wamelink & Spaapen 1999) and pharmaceutical research (dijstelbloem et al. 2002). novel modes of
quality control also figure prominently in post-normal science literature. in this context, scholars have
developed novel assessment systems which contain additional criteria in which non-academics have an
important place (van der Sluijs et al. 2005). it is unclear, however, on what scale this type of evaluations
has been adopted to date. Quality control is a broad phenomenon that comprises diverse practices. in
the nPK definition, quality control is the set of procedures and criteria that constitute the ‘selection
mechanism of problems, methods, people and results’ (p.32). This implies that it includes the assess-
ment of research proposals applying for funding, manuscripts for publication in scientific journals, appli-
cations for conference contributions, applications for academic positions, and performances of research
groups, programmes and projects. The question, then, is to what extent the novel criteria count in all
these practices. it is conceivable that relevance or applicability is of more decisive importance as a cri-
31 The new criteria employed by the funding agencies have aroused a debate including public letters signed by hundreds of university professors (Benner & Sandstrom 2000).
32 vSnu, nWo & KnaW 2003, Standard evaluation Protocol 2003-2009 for Public research organisations.
60
chapter 2
terion for attributing funding than it is for assessing candidates for academic positions. to put it simply,
the issue at stake is: What rewards do researchers receive for conducting relevant research?
The disregard of the diversity of scientific practice constitutes another weakness in the Mode 2
diagnosis. nPK raises the impression of a dichotomy of two research modes. contemporary philosophy
and sociology of science, however, emphasize the heterogeneity of scientific practices (Stengers 1997).
Scientific research is carried out in an endless variety of ways. Modern science is a ‘patchwork of very
different activities, joined together under an umbrella label, Science’ (rip 1997). it is improbable
that they can all be classified as either Mode 1 or Mode 2 knowledge production. Probably it is much
more valid to speak of Mode 1 and Mode 2 as the extremes of a continuum than of them as two mutually
exclusive categories. in this way Mode 1 and Mode 2 are ideal types, rather than really existing phe-
nomena33 and this raises the possibility to position existing practices of knowledge production on the
continuum. it seems, however, that currently the knowledge is lacking for systematically positioning the
different disciplines and subfields. although several scholars argue against the generality of nPK, they
only offer rough estimations of the differences between the disciplines.
an important lesson of our review is that in investigating changes in contemporary science systems,
one should take into account the diversity of science. due to the heterogeneity of scientific practice, the
emergence of new modes of knowledge production will not have the same impact in the whole science
system. its importance may vary in national contexts (Shinn 2002) and in scientific disciplines (al-
bert 2003). disciplinary characteristics that influence the shifts in balance between different modes of
knowledge production need not be limited to the content of their inquiries but include features of social
organization (Whitley 2000) too. further research must show how visible the various Mode 2 attributes
are in different disciplines and in different countries.
The problem of the lack of coherence is probably the largest threat to the Mode 2 concept. The
disagreement about the five attributes of Mode 2 and their relative importance shows, in the end, that
there is no compelling reason why they should operate together34. it seems more appropriate to regard
the individual attributes as separate trends than as characteristics of a general development. literature
suggests that there may be a lot of research with one or more of the features, but that the amount of work
assembling all five is marginal. for this reason, we propose to untie the wrapping around the Mode 2
concept. The individual trends that it addresses definitely deserve further research, but this should be
conducted separately, ignoring the common heading of Mode 2.
33 Muller (2000) as cited in tuunainen (2005: p. 282): nPK over-dichotomizes the evolution of science ‘presenting it as two discrete ideal types that probably never exist in their pure form in the real world’.
34 This problem is also mentioned by Steven Yearley (2005).
61
re-thinking new knowledge production: a literature review and a research agenda
To concludenPK has been successful as a manifesto. With its broad scope and evocative claims it has raised consider-
able attention in the area of science policy. it identifies a number of trends which still deserve further
consideration. a review of alternative accounts and criticisms shows that the Mode 2 diagnosis of
contemporary dynamics of scientific practice contains some adequate claims, and that some claims seem
doubtful (the rise of transdisciplinarity, reflexivity and novel modes of quality control). Moreover, the
generality of the arguments, the linear historical perspective and the necessary coherence of the original
Mode 2 arguments are all problematical. The next step we suggest in rethinking new knowledge produc-
tion is addressing the following three empirical questions:
1. do transdisciplinary research activities, with a dynamic integration of theoretical and practical com-
ponents from various disciplines, constitute a substantial part of contemporary science systems?
2. are university scientists in general increasingly reflexive, in the sense that they are aware of the
potential societal effects of their research and take these into account in their choice of research objects,
methods and approaches?
3. do new criteria, relating to the societal relevance of research results, currently count significantly in all
types of scientific quality control, not only in funding allocation, but also in retrospective evaluations of
individuals, projects or organizations?
to scholars addressing these three questions we strongly recommend taking into account the heteroge-
neity of science, paying attention to the differences between scientific fields and national contexts.
The conclusion of this paper is that the viability of an aggregate Mode 2 claim that is constituted by
five attributes is limited. our review shows that it is time to disconnect the five major constitutive claims
and to investigate them separately.
chapter 3.
63
in search of relevance: The changing contract between science and society35
Abstract This paper reflects on the relevance of academic science. relevance plays a central role in what we define
as the ‘contract’ between (academic) science and society. The manifestations of relevance in the daily
practice of academic research can be studied using the credibility cycle. together, the science-society
contract and the credibility cycle enable a systematic analysis of relevance in scientific disciplines. This is
illustrated with a case study of academic chemistry in the netherlands. We conclude that science’s search
for relevance is not new, but that its meaning changes together with changing ideas about the potential
benefits of scientific research.
3.1 Introduction
The relevance of academic research is a challenge for policy makers and science studies. in various
bodies of literature, it is claimed that the relationship between academic science and society is changing.
approaches like Post-academic Science (Ziman 2000), Mode 2 knowledge production (gibbons et al.
1994) and the triple helix of university-government-industry relations (leydesdorff & Meyer 2006) all
refer to an increasing orientation towards the production of ‘relevant’ knowledge (see chapter 2). They
observe an enhanced aspiration of academic research to contribute to the solution of societal problems
and to support innovations and economic growth. This development expresses a profound change in
the relationship between science, the state, the market and civil society. Some argue that interactions
35 This chapter was published as: hessels, laurens K., harro van lente, and ruud e.h.M. Smits. 2009. in search of relevance: the changing contract between science and society. Science and Public Policy 36 (5):387-401.
64
chapter 3
between traditional institutions are intensifying (etzkowitz & leydesdorff 2000), to such an extent that
the boundaries have become blurred (nowotny et al. 2001). due to the limits on resources available,
governments can no longer afford unconditional support for basic science (Ziman 1994) and demand
value for money, urging university scientists to demonstrate that they are effectively fulfilling their tasks.
in the meanwhile, knowledge is increasingly seen as a valuable resource to ensure economic competiti-
veness, but global competition has forced industry to cut budgets on research with a long-term horizon
(de Wit et al. 2007). They are now dependent on strategies like open innovation (chesbrough 2003),
in which universities are seen as a supplier of strategic knowledge.
it is mistaken to consider the stress on relevance as a completely new phenomenon (rip 1997).
Since the emergence of modern science, expected benefits have played a role in research funding (Mar-
tin 2003). in the course of centuries, there have been several periods in which knowledge production
was strongly connected to contexts of application (Pestre 2003). intimate interactions between science,
invention and entrepreneurship were, for example, already important in the British industrial revolution
(freeman 1997). Science is not an autarkic system; it has always depended on the provision of resources
from the government or from other wealthy individuals or organizations so it may always have been in
search of relevance. nevertheless, the question remains when and how science can be relevant. is re-
search relevant when it yields knowledge about an urgent problem such as global warming? is it relevant
when societal stakeholders are involved? is it relevant when industrial companies are willing to pay for
it? or is it relevant by definition, thanks to the cultural value of scientific activities? how is relevance
defined and measured? When and how does the stress on relevance affect scientific practices?
any attempt to address these questions should take into account that the notion of a ‘singular sci-
ence’, one model of doing and organizing science, like Mode 1 knowledge production, is historically
incorrect (Pickstone 2007). The way science is conducted at universities and the way it is influenced by
outside forces and how it reacts to societal developments has changed over time. in a broad-brush nar-
rative, Mode 1 may have shown an increase - or even a ‘lock-in’ (rip 2000) - at the expense of research
with Mode 2 features after WWii. The general conviction that basic research should be cherished like
a ‘diva’ and separated from applied research and technology development, however, is specific for that
time rather than a universal starting point for funding science and technology (Shapley & roy 1985).
The issue of relevance clearly is a complicated one. it seems that no straightforward answer is possible to
the questions above, and there is not even a common definition of societal relevance.
The aim of this paper is to contribute to develop a heuristic framework that can contribute to the
understanding of the various meanings relevance can get and the role these play in academic research
practices36. We will start with the basic notion that relevance refers to the expected value scientific
36 This paper deals only with academic science, so research conducted at universities, not at public research institutes or in industrial laboratories. for the sake of readability we will not consistently use the predicate ‘academic’ throughout the paper. everything we claim about science, scientists, research or researchers, however, is limited to university science and university scientists.
65
in search of relevance: The changing contract between science and society
research will have for society. What does this added value mean more precisely? We will investigate
how definitions of what relevance entails have shifted and may continue to shift. our argument will
proceed as follows. in the next section we will discuss the notion that the relationship between science
and society can be conceived as a contract which defines two parties and their privileges and obliga-
tions. We build on existing literature by using the idea of a contract as a heuristic device, and enrich it
by suggesting a basic composition of the contract between science and society. next, we show how an
analysis of the ‘credibility cycle’ (latour & Woolgar 1986) can help to explore the meaning and position
of relevance in actual research practices. together, our notion of the contract and the credibility cycle
constitute a heuristic framework to study ‘relevance’ in specific disciplines. in the last section we will
illustrate and substantiate the framework in a case study of chemistry in the netherlands. our first inten-
tion with developing and presenting this framework is to enhance the theoretical understanding of the
relevance of science. Second, we hope to contribute to the process of gathering empirical insights into
the changing relationship between academic science and society.
3.2 A contract between science and society
academic science is not an isolated enterprise. to keep its position in society requires legitimacy. Prac-
tices of science have changed over the course of centuries, but science has always depended on societal
support in the form of resources and legitimacy. in the development of changing identities of science,
different visions of science and connected legitimations can be discerned. early modern science could
be legitimatized by the claim that it was neutral in the sense that it would not interfere with religious
matters and it would not disturb social order. in the 18th and 19th centuries, science’s legitimacy became
dependent on the visions of science as a ‘model of rationality’ and, later, as a ‘reservoir of innovations’
(rip 1982, van den daele 1978). The large-scale institutionalization of modern science in the twentieth
century has made these latter two legitimations even more important. The substantial investments
societies make in science today are only regarded as legitimate thanks to the great promises of modern
science in terms of economic competitiveness, cultural enrichment or social progress. in other words,
the ‘relevance’ of science has become crucial for its public support. relevance can refer to practical
applications of research outcomes, but also to other ways in which science may serve society, like the
socio-cultural value of improved understanding of the world or the provision of a breeding ground for
high quality education.
numerous writings about changing science systems and concomitant legitimations employ the idea
of a (tacit) contract. as guston and Kenniston (1994b) argue the contract is a useful metaphor because:
66
chapter 3
• it ‘implies two distinct parties, each with different interests, who come together to reach a formal
agreement on some common goal’;
• a contract is negotiated, ‘arrived at through a series of exchanges in which each party tries to secure
the most advantageous terms’;
• a contract ‘suggests the possibility of conflict – or at least disparity of interests’; and
• ‘contracts can be renegotiated if conditions change for either party’ (p. 5).
Baldursson (1995), elzinga (1997), Martin (2003), Jasanoff (2005) and rip (2007) use the notion of
a contract as a heuristic for studying the changing relation between science and society. although the
approach, scope and size of their studies differ, their diagnoses are roughly similar. They start with the
‘endless frontier’ contract, which refers to vannevar Bush’s report ‘Science, the endless frontier’ (Bush
1945). it holds a clear distinction between basic research, which is self-regulated and should not be
disturbed by outside steering and applied research which is subject to immediate questions about rele-
vance and external steering. This division is institutionalized by two categories of funding agencies: basic
research councils headed by scientists and sectoral funding agencies mandated by different ministries.
for several reasons, the ‘endless frontier’ contract is currently under attack. one of the factors is sci-
ence’s own success in sectors like healthcare and agriculture, which prompts a wish to co-determine its
directions (Baldursson 1995). Society demands ‘strategic research’ (irvine & Martin 1984, rip 2004) or
‘targeted research’, new categories of basic research that combine internal scientific quality with external
societal relevance (elzinga 1997). Science-industry relationships intensify, science policy becomes stra-
tegic and funding arrangements have an increasing mission orientation (etzkowitz & leydesdorff 2000,
gibbons et al. 1994, Ziman 1994). a new contract seems to be emerging in which science’s autonomy is
increasingly constrained.
other authors use the contract notion in a more normative way. vavakova (vavakova 1998), for
instance, argues that to speak of a new contract is dangerous, because it tends to reduce the societal
usefulness of science to purely economic value. in her reading, arguments for a new contract conceive
research in a too utilitarian way. a new contract in which science is more closely linked to the private
sector would threaten the availability of knowledge as a public good. in contrast, gibbons (gibbons
1999) argues that the prevailing contract which ‘was set up to sustain the production of reliable knowl-
edge’, is not adequate anymore. given the changes gibbons observes in science and society (increasing
societal complexity, convergence of university and industrial research, rise of Mode 2), a new contract
is needed to ensure the production of ‘socially robust knowledge’ of which the validity is determined by
an extended group of experts, including lay ‘experts’. This knowledge is less likely to be contested than
knowledge which is merely ‘reliable’ in an exclusively scientific way. for gibbons, entering into the new
contract involves embracing the rise of Mode 2 and contextualized knowledge production: ‘(...) science
must leave the ivory tower and enter the agora’ (gibbons 1999).
67
in search of relevance: The changing contract between science and society
in this paper, we follow the scholars who use the contract as a heuristic but we enrich their approach
by a more specific conception of the content of the contract. here we continue the route taken by guston
(guston 2000) who defines ‘the social contract for science’ as follows: ‘The political community agrees
to provide resources to the scientific community and to allow the scientific community to retain its
decision-making mechanisms and in return expects forthcoming but unspecified technological benefits’
(p. 62). That is, science can take care of its own integrity (by self-regulation) and is able to produce
benefits for society (productivity).37
Because the contract concerns the delegation of a particular task, Principal-agent Theory (van der
Meulen 1998, Braun 2003, guston 2000) will be helpful. This approach assumes that policy-makers ask
scientists to do something for them that they cannot do themselves, because they lack the capabilities
or the knowledge the scientists have. in this view of relationships between policy-makers and scientists,
policy-makers have four fundamental problems:
• getting scientists to do what politics want (problem of responsiveness);
• being sure that they choose the best scientists (problem of adverse selection);
• being sure that scientists do their best to solve the problems and tasks delegated to them (moral
hazard); and
• knowing what to do (decision-making and priority-setting problem) (Braun 2003).
Science policy, in all its different, often complex, forms, can be understood as attempts to solve these
four problems of delegation. indeed, the evolution of science policy in the recent few decades can be
described as a sequence of different models of delegation: blind delegation, delegation by incentives,
austerity, contract delegation, and delegation to networks (Braun 2003, Potì & reale 2007).
Principal-agent Theory starts its analysis at the principal’s side and tends to overemphasize the
government’s power in structuring the relationship and to ignore the perspective of the agent (Mor-
37 for the sake of clarity let us indicate two differences between guston’s approach and ours. first, in our view science has a contract with society in the broader sense rather than with the government only. guston is mainly concerned with the government as a stakeholder in science. With his work on the ‘social contract for science’ he analyzes the changing ways in which the uS government deals with scientific research. We, however, aspire to capture the societal environment in a broader sense, taking into account all actors that have an interest in scientific research. Besides governmental authori-ties, these are all other organizations and individuals that give financial support to academic research and all that can be expected to benefit from its outcomes. in fact, one may argue, all actors in society have a stake in science, but in varying degrees.
a second difference concerns the durability of the contract. in guston’s work, ‘the social contract for science’ is a label for a specific historical period roughly from 1945 until 1980. guston uses the notion of the contract to describe and explain the particular relationship that existed between science and the government in the uSa in these years, and to contrast it with the configurations afterwards and before. in our approach, however, the meaning of the contract is not limited to a specific point in time. We argue that the idea of a contract can always be applied to the science-society relationship, regardless of the historical setting. of course the specific content of the contract can change, but its general structure will remain the same.
68
chapter 3
ris 2003). Yet, scientists also shape the institutions governing the task delegation. another limitation
is that Principal-agent Theory has difficulties with dealing with multiple principals (Shove 2003) The
framework is useful to describe the relationship between science and the government, but it can not
easily accommodate the influence of other principals like industry or ngos. in order to overcome these
limitations, we will also draw from other theories, in particular the credibility cycle (latour & Woolgar
1986, rip 1988) and resource dependency theory (oliver 1991, Pfeffer & Salancik 1978).
to conclude, the notion of a contract is a powerful metaphor and it can serve as a heuristic in order
to enhance our understanding of relevance and of the changing science-society relationship. existing
literature offers several useful starting points, but some further elaboration is necessary.
3.3 A framework to examine the relationship between academic science and society
a contract expresses the positions and the mutual relationships of the actors involved. in other words,
it draws on and expresses a moral universe, as in the case of a relation between a mother and her child,
where norms and values involved of care and protection are involved. although the norms exist and
exert force, they are not fully recorded. So, explicit and implicit norms together shape a moral universe
which indicates how mothers and children relate to each other. in the same vein, the contract between
science and society expresses implicit and explicit agreements between science and societal parties.
The contract between academic science and society deals with the delegation of a particular task.
Multiple principals ask scientific agents to do a job they cannot do themselves. from the scientists’
perspective, the contract can be regarded in terms of resource dependence (leišytė et al. 2008, oliver
1991, Pfeffer & Salancik 1978). Scientists are dependent on their societal environment for their survival,
so they cannot easily break the contract with society. if their environment develops new demands, scien-
tists can choose between several strategies. They can shift their activities towards the new societal needs,
in accordance with the ‘compliance strategy’. Sometimes they manage, however, to pursue a ‘symbolic
compliance strategy’, by creating an image which corresponds well with societal demands, without
changing anything in the actual work. alternatively they can try to negotiate with their environment
about the content of the contract, according to the ‘manipulation strategy’ (leišytė et al. 2008).
in this paper we propose to use the notion of a contract as a heuristic to study the changing
relationship between academic science and society. Just like the moral universe of the example above,
this contract cannot be found at a particular place. nevertheless, its history can be told using various
data sources like written utterances of different actors, actual policy measures, funding patterns and
interviews with key actors. assuming the existence of a contract can help to organize these data into a
coherent history of relevance. to this end, we suggest to assume the contract to contain three main com-
69
in search of relevance: The changing contract between science and society
ponents. in our conception it addresses three basic questions: what is the nature of the delegated task,
why is it delegated and how is it arranged? in other words, the contract arranges what science should
do, why it should do this, and what are the appropriate conditions for science to fulfill its delegated task
(see figure 3.1). The answers to these questions, of course, are the result of negotiations between at least
two parties. in the case of science and society there are more: neither science nor society can be seen as
one actor. Moreover, science is a subsystem of society rather than external to it. Thus, the contract is an
agreement among all actors that have a stake in academic research, including researchers doing the actual
work, university managers, policy makers and research councils facilitating it, and companies, ngos,
authorities and any other actors who are using the knowledge developed by science. The assumption of
a contract does not imply that there is no place for disagreements. The parties involved of course have
diverse interests. Still at each point in time there are a number of rather fundamental notions which the
actors do not call into question, but they refer to them as a shared vision.
Figure 3.1 Suggested composition of contract between academic science and society
The identity of scienceacademic science is concerned with conducting research in order to produce relevant outcomes.
Science’s task is to produce knowledge and to deliver it in the form of communication (publications,
presentations) or artefacts. More specifically, the contract can describe what type of research should be
carried out. What is the balance between basic and applied research, or is science supposed to conduct
strategic research (irvine & Martin 1984), or to work in Pasteur’s Quadrant (Stokes 1997)? This ques-
tion relates to the kind of knowledge science is expected to deliver, or in other words, what knowledge is
considered ‘relevant’. Societal actors may want science to offer expertise that contributes to the solution
of complex policy issues, to deliver insights in metaphysical puzzles or to produce strategic knowledge
which enhances its economic competitiveness. The contract can have (and, indeed, has had) various
concepts of relevance and these can also be specified to varying degrees. however, one can probably
assume that relevance always has a place in the identity of science, as society will demand some sort of
return for its investments.
1. What?Identity of science
3. How? Conditions
2. Why?Rationales
Science-society contract
70
chapter 3
Rationales for funding sciencein the proposed conception, the contract also describes why academic science deserves support. Many
different (combinations of) arguments are possible, but most relate to concepts of the societal relevance
of science. The expected benefits of science can take various forms, such as ‘new or improved weapons,
more accurate nautical almanacs (based on improved astronomical observations), better medical care
and agriculture, improved engines, new chemicals and materials, new energy sources, new electrical
devices and so on’ (Martin 2003 p. 19). Without claiming completeness, we mention four generic
rationales which are prominent in science studies literature. one possibility is to regard science as a
cultural good, if one believes that it has an intrinsic value or ‘axiomatic relevance’ (Spiegel-rösing 1980).
or academic research may be seen as inevitable to sustain a system of higher education (Martin 2003).
Third, one may fear that academic research is needed because the market will invest insufficiently in ba-
sic research (guston & Kenniston 1994b). finally academic science yields knowledge of general interest
(like expertise which can support governmental decision making or strategic knowledge that may lead to
medical innovations) (funtowicz & ravetz 1993, gibbons 1999, Böhme et al. 1983).
ConditionsThe third element of the contract we propose contains agreements about the conditions under which
university researchers work. ‘to create new knowledge, special procedures, norms, rewarding mechanis-
ms, and institutionalizations are put into place characterising scientific activities and distinguishing them
from other professional activities in society’ (Braun 2003). Society delegates research to science by ‘a set
of specific grants and contracts, each with its own terms and conditions’ (guston & Kenniston 1994b p.
8). for the government, the challenge is ‘how to make sure that good and useful knowledge is produced,
as well as how to assure that the investments in science do not go with unproductive pressures from the
government to produce applicable knowledge’ (van der Meulen 1998 p. 398), and the same goes for
other stakeholders of science. Science’s organization is attuned to its identity. a traditional feature of
the way academic science is currently organized is the nested hierarchy of research groups in faculties.
This may change, however, with the rise of inter-university ‘centres of excellence’ and other novel forms
of organization (rip 2004). another agreement concerns the way money is distributed to scientific
researchers, in which intermediary organizations have a central role (lepori et al. 2007, van der Meulen
& rip 1998). This is not the right place for a complete description of all agreements concerning science’s
organization. here, we just want to emphasize that the features of the organization of science in principle
aim at facilitating the fulfilment of science’s task. There can also be specific instruments (‘organizational
devices’) which aim to enhance the quality and / or relevance of research, like earmarked funding (Ben-
71
in search of relevance: The changing contract between science and society
ner & Sandstrom 2000, Potì & reale 2007) and performance assessments (geuna & Martin 2003, van
der Meulen & rip 2000)38.
as the descriptions of the three elements of the contract indicate, their precise content can change
and is subject to negotiation among all parties involved. We assume, however, that the general composi-
tion of the contract is of all times, as these three central questions are always at stake.
Credibility cycleThe next step in our investigation is to consider how ‘relevance’ manifests itself in the daily practice of
academic research. given the basic ingredients of the science-society contract (fig. 2) it is now pertinent
to ask how this may interfere in the way science operates. This undertaking can build on the sociology of
science. a central finding in the sociology of science is that scientists are not primarily driven by finan-
cial incentives and rewards, but rather by a desire for recognition and reputation. in science, ‘the need to
acquire positive reputations from other scientists is a crucial factor controlling what tasks are carried out
and how, and how they are evaluated’ (Whitley 2000). reputations play a central role in the organiza-
tion of scientific work. Scientific activities are carried out to convince fellow researchers of the impor-
tance and significance of the results and hence enhancing one’s own reputations. a scientist’s reputation
in his field is a key factor in acquiring jobs and resources. for this reason, his position in his employment
organization (university) also depends on his reputation.
Figure 3.2 The credibility cycle, adapted from latour and Woolgar (1986)
The concept of the ‘credibility cycle’ (latour & Woolgar 1986) has been introduced to explain how st-
ruggles for reputation steer the behavior of individual scientists or research groups. Similar to Whitley’s
notion of reputation, credibility refers to the ability ‘actually to do science’ (latour & Woolgar 1986 p.
198). This notion is broader than only recognition or reward for scientific achievements. it is of a more
generic nature and can take various forms such as money, data or articles. Scientific behavior, then, can
38 Most of these devices are often referred to as ‘practices of quality control’ (e.g. hemlin and rasumussen (2006)). But in the present context, this notion is a little confusing, because these devices do not only control quality, but also relevance.
Data
Money
Recognition Articles
Sta� and equipment
Arguments
72
chapter 3
be described as a cycle of conversions of different types of credibility. Scientists invest time and money
to acquire data that can support arguments. These are written down in articles, which may yield recog-
nition from colleagues. Based on this, scientists may receive new funding, from which they buy new
equipment (or hire staff) which will help to gather data again, etc. conceived in this way, the research
process can be depicted as a cycle in which conversions take place between money, staff and equipment,
data, arguments, papers, recognition, and so on (see figure 3.2). in this quasi-economic model, resear-
chers are like investors of capital who spend credibility with the intention to earn it back after a complete
cycle, with interest39.
it is important for our purposes to note that at each point in the cycle the scientific actor meets spe-
cific institutions which facilitate or constrain the conversions he is making – a point not well elaborated
in the original version of latour and Woolgar. The scientist can not make the conversions independently.
in each step there are formal or informal structures which influence the transformation of one form of
credibility in another. for example, peer review procedures govern the conversion of arguments into
articles. The criteria used by peers to assess the quality of manuscripts determine what type of argu-
ments can be converted into publications. in the next conversion, recognition is only awarded to articles
which have been accepted in prestigious journals, or – increasingly – which are cited by other scholars.
in a similar fashion, all other conversions in the cycle are also governed by institutions of various kinds,
such as university organizations, funding bodies and quality assessments. The institutions determine the
exchange rate, so to speak, of one form of credibility into the next. for the researcher, the institutional
environment is instrumental in building credibility.
two issues deserve attention here to clarify how we understand the credibility cycle. first, although
scientific researchers are the main actors in this model, it does not deal with science only. Knorr-cetina
(1982) has warned for the danger of internalism when using quasi-economic models of science such
as the credibility cycle. one should not conceive scientists as if ‘they were isolated in a self-contained,
quasi-independent system’ (Knorr-cetina 1982 p. 109). Scientists’ activities go beyond the boundaries
of their specialist community. recognition is received from peers, but also from scientific colleagues in
other disciplines. as an extreme example, nobel-prizes usually are awarded to research with an impact
extending towards other scientific disciplines or even towards products or solutions in society. What is
more, non-scientific actors play crucial roles in the acquisition of money. The availability of funding for
particular scientific projects is a product of interactions between scientists, societal actors (in industry,
government, or else), and often intermediary bodies like funding agencies. These interactions, however,
can be described with the credibility cycle, if it is not applied in a narrow sense (elzinga 1997, Wout-
ers 1999, leišytė 2007, rip 1994). We follow leišytė in assuming that credibility is achieved not only
within a researcher’s specialist community, but in ‘several arenas that interact with each other, such as
39 This does not imply that the goal of scientists is only gaining credibility and not gaining knowledge. rather, in order to take part in the research process (to gain knowledge) it is necessary to earn, retain and convert credibility.
73
in search of relevance: The changing contract between science and society
financial backers, the scientific community, regulatory authorities, and professional and consumer mar-
keting’ (leišytė 2007, p. 48).
Second, we assume that the precise composition of the credibility cycle can be subject to change. We
consider it a valuable model for scientific practice. its precise composition, however, is a historical con-
tingency rather than a necessity. What conversions are part of the cycle, which ones are most important
and what are the guiding norms and values may change over time. although latour and Woolgar (1986)
have introduced the cycle as a timeless, generic model for scientific activities, several scholars have al-
ready shown that it is useful starting point for describing changes in the science system. guston (2000),
for instance, uses the credibility cycle to explain the relationship between scientists and the government
in the uSa during the period of the ‘social contract for science’, which he positions between 1945 and
1980. Packer and Webster (1996) argue that the cycle needs to be extended in order to accommodate
patenting activities, which are of increasing importance for academic scientists. rip (1988, 1994) shows
that the credibility cycle can help to understand how the role of research councils has developed over the
years. his analysis indicates the rise of a pressure for relevance and related criteria in the assessment of
research proposals. as a result, scientists are now also involved in struggles for relevance and for legitim-
ity, on top of struggles for facticity and for fundability (rip 1988). in this paper we focus on the struggle
for relevance. We use the credibility cycle in a more specific way than rip, in a step-by-step analysis.
as a model for scientific activity, the credibility cycle can be used for studying the changing relation-
ship between science and society. The central question, then, is: how is relevance manifested in each
conversion of credibility? are there significant changes in the composition of the cycle?
A combined modelabove we have explained how the notion of a contract and the credibility cycle both can help to under-
stand what relevance is and how it figures in academic research practices. But how do both elements
relate?
The credibility describes the activities of university researchers as individuals or as organized in
groups or research institutes. in this way, it puts one actor in central focus. The conversions of the cycle
may seem primarily processes that are internal to science, conducted by the scientists themselves.
however, the institutions governing each step are co-constructed by society (to varying degrees). for
instance, the possibilities to convert recognition into money depend on the formal structures of univer-
sity funding and research councils. in turn, the criteria used in the selection of research projects to fund
are strongly determined by notions of relevance as expressed in the science-society contract. Similarly,
the production of scientific papers is in the first place ruled by peer review processes. But how scientists
act is also strongly influenced by university policy and national quality evaluations. if the allocation of
funding is increasingly based on the number of previous publications of the applicants, because this is
74
chapter 3
conceived as an indicator of quality, this will stimulate scientists to develop strategies to increase their
publication output (Weingart 2005).
in the terminology of rip (1988), scientific activity takes place at level of ‘scientific entrepreneurs
mobilizing resources’, but the intermediary level ‘organized field of scientific institutions’ also give shape
to the credibility cycle. Some of the credibility conversions are ruled by institutions on the intermedi-
ary level, such as research councils or disciplinary associations. These institutions constrain actions and
interactions of scientists, but also enable them by providing legitimacy and opportunities. The science-
society contract can be located on the macro level of the ‘general social context’ (rip 1988) in the sense
that it is of a general kind and all stakeholders of science take part in it.
The science-society contract constitutes a platform for all actors in the research system which
provides them with arguments, legitimacy and other resources. The particular specification of ‘relevance
of science’ in the operative contract precipitates into the intermediary level organizations and therefore
shapes the credibility cycle. The conditions which are generally believed to be needed for science to
optimally fulfill its task are translated into institutions such as university organizations and research
evaluations.
from the scientist’s perspective, the conversions of the credibility cycle may appear as exchanges
of different currencies. all forms of credibility constitute pools of resources that can be transformed
into one another. The exchange rate of these conversions, however, is determined by the contract. for
instance, the criteria ruling the selection of manuscripts for publication, the selection of candidates for
academic positions, or the selection of research proposals for funding depend on the content of the
contract between science and society. The agreements about the conditions probably have direct influ-
ence on credibility conversions, while the identity of science and the rationales for public support will
indirectly contribute to the institutions steering the conversions of credibility.
in their turn, developments in scientific practice also contribute to shifts in the content of the
science-society contract. as actors refer to concepts of relevance in their daily work, they reinforce
the contract. in giving credit to particular research contributions or in deciding how to spend money,
scientists refer to shared views on what science is and what it should or should not contribute to society.
for instance, technological expectations that go together with research outcomes influence ideas about
potential benefits of a particular field for society (van lente 1993, van lente & rip 1998). Scientific re-
search yields insights both in problems to be solved (e.g. climate change) and in possible solutions (e.g.
metal hydrides for hydrogen storage on board of vehicles) (Bakker et al. 2008). These promises can give
rise to dedicated research programmes, schools or institutes on the meso-level that subsequently influ-
ence the negotiations about the science-society contract. expectations about how particular science and
technology can solve societal problems shape concepts of the societal relevance of science in general,
too. Particular expectations may deal with specific domains, but in the end the overall view on the value
of science will be affected by the expectations in all fields together.
75
in search of relevance: The changing contract between science and society
to conclude, the contract enables and constrains scientists and others in their activities. it expresses
the agreements among all actors and determines the rules and regulations of the institutions which give
shape to the credibility cycle. in this way, it makes particular conversions of credibility possible, but
complicates others. There is no uni-directional causality between the contract and the various credibility
cycles. The contract influences academic practice, and developments in practice, in their turn, influence
the agreements in the contract. Because of these interactions, we believe that a study of the relevance
of science requires analyzing the contract as well as scientific practice. in the following section we will
illustrate this in an empirical case study.
3.4 Struggles for relevance in academic chemistry
We now go one step further, and investigate how concepts of relevance and the struggles around it are
visible in a case study. as a unit of analysis, the scientific discipline seems appropriate. in this section we
narrow down the focus from science in general to a case study of academic chemistry in the nether-
lands40. obviously the empirical outcomes of this case study do not represent science in general. The
meaning of relevance can be different in each field of research. an interesting aspect of chemistry is the
intensity which the relationships between universities and industry traditionally have. for this reason,
one can expect the relevance of chemistry to be strongly connected with industrial applications and
the relationship between science and society to be relatively stable. as such it seems a ‘hard case’ for
studying changing notions of relevance and an interesting test for our framework. given the heterogen-
eity of science, this case can obviously not be taken as representative of the whole science system. in
some other fields, more radical transformations may occur, especially if considerations of economic or
commercial relevance are traditionally absent. our analysis starts with the post-war situation, previously
referred to as the era of ‘Science, the endless frontier’. first we will describe the changes in the identity,
the rationales and the conditions that have regulated academic chemistry research in the netherlands.
Then we will address the changing credibility cycle.
Identityin the first postwar decades, some professors have strong ties with chemical companies, but the main
task of academic chemistry is to conduct basic research not directly aiming at industrial applications.
The relevance of academic research is often defined in terms of a contribution to chemical industry, but
more in terms of skilful and knowledgable r&d staff than in terms of useful knowledge.. Knowledge
40 This 10 month case study is based on document analysis and in-depth interviews with scientists, policy-makers and other societal stakeholders. here we summarize our findings, which requires simplifying a complex history involving various actors with different stakes and interests. a more elaborated analysis of this case will be published separately.
76
chapter 3
transfer to industry takes place through the people who move to industry after their (doctoral) gra-
duation, by consultancy services and by industrially funded research assistants (homburg 2003).The
usefulness of academic chemistry is not called into question, given the importance of chemical industry
for the dutch economy and the linear model-based belief that basic research will ultimately lead to
innovation and economic growth. in the 1970s, when “science policy” is introduced, the government
demands academic research to more effectively address societal needs41. The minister of science policy
emphasizes that chemistry should produce knowledge which is useful not only in chemical industry,
but also in domains like agriculture and healthcare. in the 1980s, when innovation emerges as a central
topic for policy making, chemistry’s relevance is broadened to include delivering applicable knowledge,
too. The main activity remains basic research, but this is justified merely as a necessary condition for the
existence of good applicable research. during the 1990s the idea of ‘strategic research’ emerges which
blurs the boundary between basic and applied research. chemistry’s task then becomes producing
‘strategic knowledge’ which combines aspects of basic and applied research. Strategic knowledge can be
defined as fundamental insights in domains of high relevance for economy or society (irvine & Martin
1984, rip 2004). The domain of ‘sustainability’, which gains importance since its introduction in 1987,
becomes a major strategic field for chemistry42. This notion combines the aspirations to contribute to
economical growth and, at the same, to solve environmental issues. The strategic identity of academic
chemistry endures in the new millennium as it is compatible with the most recent innovation concepts,
in which the university is seen as a supplier of basic knowledge which can be valorized by other actors in
the innovation system.
RationalesThe most dominant rationales for funding academic chemistry of the 1950s and 1960s seem to be its
necessity for the training of new r&d-workers and (less importantly) the cultural value of basic research
(see table 3.1). chemical industry, with which academic chemistry always had strong bonds (hom-
burg & Palm 2004a), has a strong interest in highly educated researchers. although some university
researchers also provide industry with technical advice, in the post-war period, industry tends to regard
universities primarily as an educational system to supply a steady stream of highly skilled researchers
(homburg 2003). The research council Son, founded in 1956, provides support to basic chemical
research with the training of young scientists and the stimulation of cooperation as its primary aims. in
the 1970s, when governmental science policy is born, the need for chemical research is framed in more
general terms, going beyond its value for chemical industry. in justifications for its public support the po-
41 in the first policy paper (1974) by the first dutch minister of Science Policy, Mr. trip, the primary mission of science policy is defined as enhancing the agreements of research agenda’s with societal demands.
42 foresight studies by ocv (ocv 1995) and by Kncv and vnci (Kncv & vnci 1994) both pay significantly more attention to environmental issues than their predecessors from the 1980s.
77
in search of relevance: The changing contract between science and society
tential to contribute to problem-solving in other sectors gains visibility43. This broadening is a response
to the negative image of the chemical sector, due to the rising awareness of environmental problems.
The budget cuts on basic research in industry in the 1980s increase the importance of the rationale for
supporting academic chemistry relating to its potential contribution to technological innovations44.
chemical industry does not only desire to benefit from academic research in terms of educated staff,
but also in terms of applications of the knowledge produced45 (de Wit et al. 2007, van helvoort 2005).
in the 1990s the notion of sustainable development becomes increasingly significant in rationales for
funding (chemical) research46. This umbrella concept connects several previous rationales because it
refers to growth in economic and environmental dimensions47 (rip 1997, van lente & van til 2008).
around the turn of the century, the funding of university research is increasingly framed as support for
the national innovation system. Maintaining a “healthy innovation system” becomes a central goal of
economic policy, which implies that universities deserve support thanks to their central position in this
system. in this perspective, however, support for university researchers is accompanied with the expecta-
tion that they actively interact with other actors in the innovation system, and contribute to the process
of “valorization”, by writing patents or by founding spin-off companies.
Conditionsduring the 1950s and 1960s chemical scientists have a high degree of autonomy. The most important
types of funding (university funding and research council Son) are distributed without any conditions
attached, based on considerations of academic quality. industry also pays for research assistants and
consultancy service, but only to a minority of all professors (homburg 2003). in the 1970s scientists are
increasingly held accountable for their work. although they are not yet affected by formal policy mea-
sures, chemical researchers are under pressure to put more effort in explaining their research priorities
to policy makers and funding agencies. from 1980 onwards, however, a substantial change occurs in the
43 e.g. nota Wetenschapsbeleid 1974.44 The first foresight study (1980) of chemistry, commissioned by the minister of science policy concludes that academic
chemistry should define its research goals more sharply and that the contacts with industry deserve intensification. 45 The interest of chemical industry in academic research is also visible in an advisory report by vnci (association of
chemical industry) and Kncv (professional organization of chemists) that argues for increasing industrial steering of academic research and suggests a set of six ‘priority fields’ as a guideline in the conditional funding process.
46 ‘for all new scientific and industrial activities on the field of chemistry the strives for sustainability and the mini-malisation of environmental pressure have become important boundary conditions.’ (original emphasis) (ocv 1995, chemie in Perspectief: een verkenning van vraag en aanbod in het chemisch onderzoek, overlegcommissie verkenningen, amsterdam). nWo’s Strategy note on chemistry for 2002-2005 is even titled: ‘chemistry, Sustaina-ble and interwoven’ (nWo cW 2001, chemische Wetenschappen Strategienota 2002-2005: chemie, duurzaam en verweven, nederlandse organisatie voor Wetenschappelijk onderzoek, den haag).
47 The 1995 foresight study on academic chemistry, for example, writes about sustainability: ‘Much of the research on re-sources (...) is not only motivated by economic considerations but serves at the same time important ecological goals.’ (ocv 1995, chemie in Perspectief: een verkenning van vraag en aanbod in het chemisch onderzoek, overlegcommis-sie verkenningen, amsterdam).
78
chapter 3
funding of chemical science, due to two interrelated developments. first, faced with the need to restrict
the growth of science funding (Ziman 1994), the ministry of Science, culture and education starts to
supply part of the first money stream on a conditional base (Blume & Spaapen 1988). Moreover, in-
formed by foresight studies48, an increasing share of the second money stream is dedicated to application
oriented research49. Second, a growing part of all funding stems from industrial corporations50 and the
ministry of economic affairs starts to implement a number of innovation programmes in specific fields
like carbohydrates and catalysis51. These developments continue in the 1990s. The reorganized research
council nWo continues to broaden its mission beyond basic research and increasingly works with
‘priority programmes’ and ‘fields of attention’ that are chosen partly thanks to their societal relevance52.
another significant event in the 1990s is the institutionalization of performance evaluations. govern-
mental policy makers, research councils and university managers who have become more selective in
supporting particular activities, develop a need for transparency with regard to research outcomes. in
1996 the first nation-wide quality assessment of chemical science is conducted53, the second in 200254.
due to the lack of a strict protocol, the evaluators can choose themselves to what extent they take into
account considerations of societal relevance as for instance the economic value or technological applica-
tions of the produced knowledge (van der Meulen 2008). in practice, they turn out to generally ignore
this type of criteria and focus strongly on traditional scientific norms55. after 2000 chemistry faces a
further diversification of policy instruments. Thanks to their continued growth, the european frame-
work Programmes become a substantial source of income for academic chemists. Moreover, there is a
rise of consortia-based funding, large sums of governmental money supplied to collaborative programs
of university scientists which are monitored by (industrial) user committees and which explicitly aim at
enhancing the interactions with industry56.
48 chemie, nu en straks: een verkenning van het door de overheid gefinancierde chemisch onderzoek in nederland, 1980, Staatsuitgeverij, den haag; over leven: Betekenis van de Biochemie in nederland, 1982, Staatsuitgeverij, den haag
49 Son, the major chemical research council, starts a program for applied chemical research in 1980, together with the new technology foundation StW. Moreover, Son’s mother organisation ZWo (‘organization for Pure Scientific re-search’) is reorganized in 1988 into nWo (‘dutch organization for Scientific research’), in which there is more room for funding application oriented research.
50 during the 1980s, the number of temporarily paid chemical researchers funded by the ‘third money stream’ rises from about 70 to 350 (acc-evaluatiecommissie 1991, evaluatie van de universitaire chemie in de jaren ‘80, acc/KnaW, amsterdam).
51 The ‘innovation oriented Programs’ (ioPs), for example ‘Membranes’ (1983), ‘carbohydrates’ (1985) and ‘catalysis’ (1989).
52 Son Jaarverslag 1991, 1993, 1995.53 vSnu 1996, Quality assessment of research: chemistry: past performances and future perspectives. 54 vSnu 2002, assessment of research Quality: chemistry and chemical engeneering. 55 This may change, as in the meanwhile a protocol is available (vSnu et al. 2003) and there are attempts to develop
more holistic methods that include societal relevance (Spaapen et al. 2007)56 e.g. the actS-program (advanced chemical technologies for Sustainability) of nWo, the technological top insti-
tutes funded by the ministry of economic affairs and the BSiK-programs (Besluit Subsidies investeringen Kennisinfra-structuur).
79
in search of relevance: The changing contract between science and society
Table 3.1 overview of the changing contract for academic chemistry. + signs indicate that these elements complement existing elements rather than replacing them
Summary of identity Most dominant rationales Most important conditions
50s and 60s
basic research educationcultural value
autonomyunconditional fundingSon communities
70s + useful knowledge + problem solving potential + social accountability
80s applicable knowledge technological innovation + conditional funding+ application oriented funding (StW, ioP, contract research)+ foresight+ scarcity of resources reorganization nWo
90s strategic knowledge +sustainable development further prioritization+ performance assessments
2000+ strategic knowledge + innovation system + consortia (actS, TTi, bsik)+ european fPs
to summarize, there have always been bonds between academic chemistry and industry but the type of
interaction has changed. The meaning of relevance has changed in the course of years. initially education
and cultural value ruled its definition; later serving society and the environment; in the 1980s innova-
tion became dominant; since the 1990s specified in terms of sustainability. related, the emphasis in the
rationales for funding chemical research have shifted from its function to support higher education and
its cultural value to the notion that basic research is needed to sustain the innovativeness of industry sin-
ce global markets fail to stimulate private sector basic research. an additional rationale that has evolved
over the years is the need of chemical expertise for governmental decision making about regulations of
emissions. The conditions specified in the contract have become increasingly complex. chemists receive
less unconditional support. Still the ministry of science provides a certain share of funding without spe-
cifying how it should be spent, but the degrees of freedom in spending this ‘basic funding’ is decreasing
too57. Moreover, for a fruitful career, scientists depend on the acquisition of additional funding, from
nWo, european framework Programmes or from private companies. each of these sources has speci-
fied targets and requires from researchers to define ex ante the societal significance of the research they
propose. Moreover, a couple of new devices are in place to stimulate the production of good and relevant
knowledge: performance assessments and foresight activities.
57 due to university management like the current ‘focus en Massa’ policy (e.g. universiteit utrecht, 2005, Strategisch Plan 2005-2009, utrecht) and due the requirement of ‘matching’ externally funded research projects with block-grant support (aWt 2004).
80
chapter 3
Credibility cyclechanges in the identity, rationale and conditions of academic chemistry will have an impact on scientific
practice, which can be analysed in terms of the credibility cycle. The institutions around each conversion
in the cycle are influenced by changes in the contract. Some conversions seem solely ruled by the scien-
tific community, but in other cases external parties deliberately interfere. in our case study we followed a
number of ‘organizational devices’ that have been designed to enhance a particular form of relevance of
scientific research. in the case study, we identified five types of these devices:
• earmarked funding;
• foresight activities (e.g. verkenningscommissies, Sectorraden);
• internal (scientific) procedures of quality control (peer review of scientific papers, selection of can-
didates for academic positions, citation practices);
• university management, (e.g. ‘focus and mass’-policy, promotion criteria); and
• performance assessments (visitaties)
in the following we will discuss how these organizational devices interfere with particular credibility
conversions (see figure 3.3) and, thus, how ‘relevance’ has been expressed in the practices of dutch
academic chemistry research.
Figure 3.3 The credibility cycle, adapted from latour and Woolgar (1986). Points at which organizational devices connect to the cycle are shown.
peer review
earmarked funding
selection procedures
foresight
Data
Money
Recognition Articles
performance assessment (and/or citations)
(and other outcomes)
Sta� and equipment
1
6
5
4
3
2
Arguments
81
in search of relevance: The changing contract between science and society
1. first, the rise of earmarked funding interferes with the conversion from recognition to money. The
criteria connected to particular funds determine the possibilities for scientists to obtain money. in this
way they co-shape the prevailing norms that govern this specific conversion. The availability of a number
of strategic programmes for chemical research in the netherlands implies that scientists who are recogni-
zed experts in particular fields (like catalysis or nano-science) have an advantage in acquiring money.
also the ability to promise practical applications is increasingly valued in this conversion. Moreover,
earmarked funding is indirectly influenced by foresight activities. from a principal-agent perspective,
foresight studies can be seen as an attempt of a principal to overcome delegation problems, since the
studies seek to make the world of the agents more transparent to the principal (van der Meulen 1992).
The foresight studies on chemistry58 in the netherlands have had a significant influence on the available
funding for particular sub-fields59.
2. The procedures for selecting candidates for academic positions influence the conversion of money
into staff (and equipment). in these procedures normally a combination of various criteria is taken
into account. interviews with scientists indicate that recently, the publication and citation records have
gained importance here. The most recent trend is that a candidate’s proven abilities in funding acquisi-
tion are rewarded. obviously, these developments co-determine how money can be transformed into
staff.
3. The conversion of staff and equipment into data takes place relatively autonomously inside laborato-
ries and other research locations. governing norms seem to be primarily of scientific nature. however,
the choice of what data to produce can strongly depend on agreements with funding sources and – less
explicitly – on societal issues calling for particular scientific knowledge. The frequency of interaction
with potential users during the research process varies across funding sources, from about two to six
times a year. however, in interviews academic researchers speak about attuning their activities to user
needs as a boundary condition rather than a primary goal.
4. The conversion of data into arguments depends on norms about the characteristics data should
possess before they can support arguments: reliability, validity. traditionally, there is little societal
interference here. Proponents of post-normal science (funtowicz & ravetz 1993) claim this to change,
the public becoming increasingly involved in quality assessments of scientific results in fields of social
importance, but we have not observed this in our case study. Some interviewees, however, indicate that
58 chemie, nu en straks: een verkenning van het door de overheid gefinancierde chemisch onderzoek in nederland, 1980, Staatsuitgeverij, den haag; ocv 1995, chemie in Perspectief: een verkenning van vraag en aanbod in het chemisch onderzoek, overlegcommissie verkenningen, amsterdam.
59 for example, Son’s decision to start funding application oriented research is partly based on the outcomes of the 1980 foresight study (Son Jaarverslag 1980).
82
chapter 3
sponsoring organizations sometimes attempt to influence the formulation of arguments because they
have a strong interest in particular outcomes, for example regarding the safety of chemicals.
5. The conversion of arguments into articles is strongly ruled by peer review. all scientific journals
employ this practice of quality control. Peers judge the quality of papers in academic terms, without
considering their content’s societal relevance. a significant modification of this conversion concerns
the rising interest of research sponsors, particularly in industrial fields like catalysis, to patent research
outcomes. for this reason industrial funding often implies a delay of 1-3 months for publications, due to
the scientists’ contractual duty to give their industrial partners some time to explore the patentability of
their findings. We also found that chemical researchers are actively involved in writing patents, especially
in the field of catalysis.
6. obtaining recognition based on one’s articles and patents traditionally depends on peer judgment of
innovativeness and contribution. however, in the 1980s and 1990s the government and the union of
universities have installed performance assessments that formalize the attribution of recognition (van
der Meulen 2008). Bibliometric analyses combined with qualitative judgments from peers (and some-
times also of ‘knowledge-users’) yield scores for the performance of research groups, programmes and
individuals. interview data suggest that these scores increasingly contribute to an individual’s scientific
recognition. although they do not play a dominant role yet, the intended involvement of ‘users’ and
including the ‘societal relevance’ as a significant criterion in these assessments (Spaapen et al. 2007) may
considerably modify the conversion of articles into recognition.
note, however, the paradox here. Performance assessments, including bibliometric analyses, have
increased the ‘publish or perish’ norm (Weingart 2005, Wouters 1999). although they were started to
enhance the societal accountability of scientists (van der Meulen & rip 2000), they have increased the
need for peer recognition rather than the need for societal justification. The development of bibliome-
tric evaluation tools has added a quantitative dimension to the conversion from articles to recognition.
originally a means of communication, publications have become an end in itself.
The new mosaic of funding options, partly influenced by foresight activities, cause that only a happy
few manage to get money without promising some sort of practical applications. in the contract of the
1950s and 1960s, scientists had many options for acquiring resources based solely on considerations of
originality or scientific relevance, but this has become a rarity nowadays. one effect of the increasing
complexity of funding options is that acquisition activities today consume a major share of senior scien-
tists’ time. This development has also influenced the procedures for selecting candidates for academic
positions. Scientific publication records still seem to be the major criterion here, but increasingly the
83
in search of relevance: The changing contract between science and society
candidate’s estimated fundraising abilities60 are rewarded, too. together the changed conditions under
which scientists work have strengthened the mutual competition. The pressure to publish and to be cited
has intensified61.
3.5 Conclusion
The framework we have developed in this paper enriches the understanding of how scientific practice
relates to external pressures and how internal developments influence science’s relation with society.
today, relevance is at the core of the relationship between academic science and society. The legitimacy
of public support for university research depends on its perceived potential benefits, ranging from
contributions to culture or education to specific insights or products with economic value. The idea of
a contract as a negotiated set of rights and obligations to understand this relationship turns out to be
helpful. after making assumptions about the composition of this contract it can serve as a heuristic.
continuing on the route taken by guston (2000), in this paper we have attained this higher degree of
specificity, by conceptualizing the general content of a science-society contract in terms of ‘identity’, ‘ra-
tionales’ and ‘conditions’. Moreover, we have created a link with the credibility cycle (latour & Woolgar
1986), turning it into a tool to study the role and position of relevance in academic practices.
The ‘relation between science and society’ and ‘scientific practice’ both are vague concepts trying to
grasp complex pieces of reality. The strength of our approach is that we have developed a model in which
both are reduced to something that does justice to reality and at the same time allows for systematic
analysis. our notion of the contract clearly articulates the relation between science and society in terms
of a limited number of variables. Similarly, the credibility cycle may not give an exhaustive overview of
all activities involved in scientific research but it does describe the general pattern in which all activities
have a position. assuming that all scientists’ actions aim to contribute to the conversion of credibility,
the form of this cycle becomes a crucial factor determining scientific practice. our framework does not
imply any form of determinism. The cycle is not a result of only the contract, but of social interactions
within and outside science. in the same way, the credibility cycle is not a dictating norm system, but
rather a particular pattern of activities that are necessary for each researcher to produce credible informa-
tion. ‘norms, the socialisation process, deviance, and reward are the consequences of social activities
rather than its causes’ (latour & Woolgar 1986, p. 205).
60 important indicators for the assessment of a candidate’s fundraising ability are his / her past funding acquisitions and networking skills (evidence from interviews with scientists).
61 This is confirmed in our interviews with chemical researchers: ‘Without it i do not survive’ (full professor in biochem-istry) and ‘i think it is the only way to show what you have done. and i think that if something is not publishable in a scientific journal, it is not worth much’ (Phd-student in biochemistry).
84
chapter 3
We believe to have demonstrated the usefulness of our framework in a case study of academic chem-
istry in the netherlands. The framework helped us to identify the ways in which concepts of relevance
change and how they influence scientific practice. in particular, we have distinguished various ‘organi-
zational devices’, which aim to enhance the relevance of scientific research, the most influential ones
being earmarked funding and performance assessments. an important finding in our case study is that
the effects of the changes in the credibility cycle are partially in contradiction, creating a paradox around
relevance. The pressure for strategic knowledge is most visible at the conversion from recognition to
money, but at most other conversions it is of little importance. on the contrary, in the other conversions,
the importance of academic criteria of success has increased further due to the rise of bibliometric qual-
ity criteria. in some subfields of chemistry this creates a tension between different parts of the cycle. This
tension is absent in catalysis, where the applicability of basic research is easily demonstrable. But in other
fields, like biochemistry, there is a difference between research which is excellent according to scientific
peers and research which is considered useful by societal actors. This implies that the criteria guiding the
selection of research proposals contradict the criteria guiding the selection of manuscripts for publica-
tion or the attribution of recognition. Scientists are then rewarded for making beautiful promises about
the (possible) relevance of their research, but not for realizing these promises.
Possible next steps are to elaborate this case and to extend the approach to different scientific
disciplines and to other national contexts. The way our framework helped us to observe changes in the
relationship between academic chemistry and society suggests that it will work for other cases, too. Most
other disciplines do not traditionally have such a strong relationship with one industrial sector, so they
may show even stronger developments.
it is often assumed that the pressure for relevance has increased during the past few decades. how-
ever, it seems more fruitful to conclude that relevance has got different meanings and different forms
in which it is expressed and expected. These forms can also be in contradiction, as our case study has
shown. The meaning of relevance changes together with changing ideas about the potential benefits of
scientific research. here we have introduced a framework that facilitates a systematic empirical study of
changing concepts of relevance as products of the interaction between science and society.
at least two implications for public policy follow from this paper. first, an enhanced understanding
of the changing societal position of university research is crucial for making sound policy in the field of
science and innovation. This is often based on popular and unarticulated notions of societal relevance,
without a clear understanding of what these entail. This paper shows that it is important to be careful
with the term ‘relevance’, as its meaning is variable in time and place. The notion sometimes even is an
empty shell, as actors use it to their own benefit. Second, our brief case study of academic chemistry in-
dicates that the effect of public policy is limited when it only intervenes at one position of the credibility
cycle. Policy aiming to enhance the correspondence between academic research agendas and societal
85
in search of relevance: The changing contract between science and society
needs would be more effective if a more systemic approach is taken which includes a combination of
instruments directed at various conversions of credibility.
chapter 4.
87
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry62
Abstractin many western science systems, funding structures increasingly stimulate academic research to con-
tribute to practical applications, but at the same time the rise of bibliometric performance assessments
has strengthened the pressure on academics to conduct excellent basic research that can be published
in scholarly literature. We analyze the interplay between these two developments in a set of three case
studies of fields of chemistry in the netherlands. first we describe how the conditions under which
academic chemists work have changed since 1975. Second we investigate whether practical applica-
tions have become a source of credibility for individual researchers. indeed, this turns out to be the case
in catalysis, where connecting with industrial applications helps in many steps of the credibility cycle.
Practical applications yield much less credibility in environmental chemistry, where application oriented
research agendas help to acquire funding, but not to publish prestigious papers or to earn peer recogni-
tion. in biochemistry practical applications hardly help in gaining credibility, as this field is still strongly
oriented at fundamental questions. The differences between the fields can be explained by the presence
or absence of powerful upstream end-users, who can afford to invest in academic research that promises
long term benefits.
62 This chapter was co-authored with harro van lente and has been submitted for publication.
88
chapter 4
4.1 Introduction
under labels such as ‘entrepreneurial science’ (etzkowitz 1998), Post-academic Science (Ziman 2000),
and Mode 2 knowledge production (nowotny et al. 2001, gibbons et al. 1994), influential scholars have
reported an increasing intertwinement of university research with practical applications. however, these
diagnoses have been criticized for their theoretical shortcomings and for a lack of empirical support
(Pestre 2003 , see also chapter 2). our starting point in this paper is that two major developments can
be discerned in the governance of academic research, which may be (partly) in contradiction. first,
the pressure on academic research has grown to contribute to practical applications of the knowledge
it produces. Public support for university research has shifted from block-grant support to earmarked
funding for specific projects and programs (Morris 2000, lepori et al. 2007). university researchers are
increasingly stimulated to engage in for-profit activities, patenting and subsequent royalty and licen-
sing agreements, spin-off companies and university-industry partnerships (Slaughter & leslie 1997,
geuna & nesta 2006). at the same time, however, the rise of quantitative performance evaluations has
increased the need for scientific accountability, which could enhance the pressure for publications in
academic journals (hicks 2009, Wouters 1997). While it seems of crucial importance for the future of
academic science, the interplay between these two developments has received little attention so far. The
current paper aims to fill this gap. We explore how the shifts in funding and the rise of evaluations have
taken place in a specific discipline and analyze how these developments shape the research practices of
individual researchers, in particular their orientation towards practical applications.
The central question of this paper is: have the changes in the science-society relationship made
practical applications into a source of credibility for academic scientists in three fields of chemistry?
our analysis consists of two major steps. first we give a detailed analysis of the changing relationship
between dutch academic chemistry and society using the framework of a science-society contract. Second
we systematically analyze the role of practical applications in the research practices of three fields of
chemistry.
We have chosen for a comparative case study set-up, to be able to adequately deal with the heteroge-
neity of science. labels like entrepreneurial science or ‘Mode 2’ tend to obscure the diversity of science,
treating it as a monolithic system moving from one state to the other (heimeriks et al. 2008, see also
chapter 2 of this thesis). however, changes in the governance of science will have different implications
for different research fields, due to their cognitive and organizational differences (Bonaccorsi 2008,
Whitley 2000, albert 2003). Moreover, national science policies, the presence of particular industrial
sectors and cultural variation considerably influence the dynamics of science-society relationships (rip
& van der Meulen 1996, Shinn 2002).
89
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
4.2 Theoretical framework
our theoretical starting point is that scientists and their organizations are no isolated entities, but
they interact with their environments to achieve their objectives. They depend on their environments
for critical resources like funding and legitimacy (leišytė et al. 2008, Pfeffer & Salancik 1978). from
this perspective, on the macro-level the relationship of academic research with society can be seen as
a ‘contract’ (Martin 2003, elzinga 1997, guston & Kenniston 1994a). in this paper, we build on the
concept of the science-society contract developed in chapter 3 (hessels et al. 2009). This contract is not
a physical entity, but a representation of the moral positions that encompasses all implicit and explicit
agreements between academic science and governmental departments, ngos, firms and other societal
parties, specifying what science should do (identity), why it should do this (rationale), and the appro-
priate conditions for science to function well (conditions), see figure 4.1. according to this contractual
perspective, the very identity of science is connected to the provision of a valuable public good: relevant
research outcomes. Science’s task is to produce knowledge and to deliver it in forms like papers, patents,
artifacts or educated people. The precise type of expected knowledge (basic, applied) and the degree
to which science should be involved with practical applications vary over time and across disciplines.
The contract, that is, the set of implicit and explicit agreements, also describes why science deserves
support. academic research is often regarded as a necessary stipulation for sustaining a system of higher
education, commercial product development, and informing complex decisions and innovation. The
third element of the contract, which will receive most attention in this paper, contains agreements about
the conditions under which scientists work, including expectations regarding the social structure of the
research community, allocation of research funds, and incentives for producing more socially relevant
knowledge. The changing science-society contract of dutch chemistry will be analyzed in Section 4.4.
The resource dependency of individual researchers, on the other hand, can be expressed by the
‘credibility cycle’ (latour & Woolgar 1986). This model explains how struggles for reputation steer the
behavior of individual scientists. its starting assumption, underpinned by many sociological studies of
science, is that a major motivation for a scientist’s actions is the quest for credibility. Scientists invest
time and money expecting to acquire data that can support arguments. These are written down in
articles, which may yield recognition from colleagues. Based on this, scientists hope to be able to receive
new funding, from which they buy new equipment (or hire staff) which will help to gather data again,
etc. conceived in this way, the research process can be depicted as a repetitive cycle in which conver-
sions take place between money, data, prestige, credentials, problem areas, argument, papers, and so on.
in Section 4.5 we analyze the role of practical applications in the credibility cycle of three fields of dutch
chemistry.
90
chapter 4
Figure 4.1 our framework for studying the changing relevance of academic chemistry (based on chapter 3 (hessels et al. 2009))
4.3 Methods: a case study of chemistry in the Netherlands
for our exploration of the effects of institutional changes on academic research practices, we use a set of
case studies of three chemical fields in The netherlands63. investigating the potential tension between
pressures for academic publications and pressures for practical impact requires a discipline with both a
strong publication tradition and possibilities to turn research outcomes into practical applications. che-
mistry fulfills both conditions well. in dutch chemistry there is a long tradition of relationships between
university researchers and companies (rip 1997, homburg 2003). There is an exceptionally cooperative
relationship between universities and industry in this field, also in the form of collective lobbying for
public money (a07, r06). Moreover, chemistry has a strong academic tradition, and dutch chemists
have an excellent reputation for their scientific publications (Moed & hesselink 1996).
Still, connections with industry are not of the same intensity across all chemical fields. Because we
aim to explore the diversity of science, our case studies deal with three fields with different relation-
ships to industry and other societal stakeholders. Biochemistry is a relatively fundamental field and has
relatively few interactions with societal organizations. its main applications are in the medical domain
and are mediated by medical researchers. environmental chemistry contributes directly and indirectly to
environmental policy. it also delivers knowledge and tools to industry and non-governmental organiza-
tions related to the risk assessment of industrial chemicals. catalysis is strongly connected to chemical
63 These case studies have also been used as a brief empirical illustration in an earlier, more theoretical paper (hessels et al., 2009). The current paper presents a more detailed analysis of the material, focusing in particular on the differences between the three fields.
Contract between science and society
Money
Identity
Conditions Rationales
Data
Recognition
Sta and equipment
Credibility cycle
Articles
Arguments
(and other outcomes)
91
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
industry. its knowledge can help firms to enhance the efficiency of their production processes and to
decrease their environmental impact.
our analysis of the changing science-society contract of dutch chemistry (Section 4) is based on the
documents listed in appendix a in combination with interviews with four scholarly experts64 on dutch
chemistry, r&d officers of five chemical companies, and representatives of research council nWo,
the association of dutch chemical industry (vnci), the dutch chemical association (Kncv) and the
department of science policy of the Ministry of education, culture and Science. The documents were
collected based on prior knowledge of the authors, tips from interviewees, and the ‘snowball method’.
The selection includes governmental policy documents, reports and strategic plans of research coun-
cils, foresight studies, evaluations and other important publications about dutch academic chemistry.
The findings from these documents were triangulated in interviews with the experts and stakeholders
mentioned above. in this paper we will refer to these documents using the abbreviations presented in the
appendix. The contract analysis is delimited to the period of 1975 until about 2005. The starting year of
1975 marks the beginning of governmental science policy in the netherlands (M74), which is generally
regarded as a landmark event in the growing societal demand for application oriented research.
for the credibility cycle analysis (section 4.5) we carried out semi-structured in-depth interviews
with 20 academic researchers in catalysis, environmental chemistry and biochemistry. The respondents’
ranks ranged from Phd-student to full professor and they were employed at five different universities in
the netherlands (see table 4.1). We asked them questions about their current and past research activi-
ties, their personal motivation, and their experiences and strategies concerning funding acquisition, pub-
lishing, scientific reputation, and performance evaluations. using nvivo (qualitative analysis software),
the interview transcripts were coded in accordance with the different steps of the credibility cycle. in the
interview analysis special attention was given to differences among the three scientific fields.
Table 4.1 distribution of respondents over fields, universities and academic ranks
catalysis (9) Biochemistry (6)environmental chemistry (5)
university of amsterdam (8)utrecht university (6)vu university amsterdam (3) radboud university nijmegen (1) eindhoven university of technology (1)leiden university (1)
full professor (6)retired full professor (5)associate professor (5) Phd-student (3) Post-doc researcher (1)
64 Scholars in the field of Science, technology and innovation Studies with expertise on chemistry: prof. dr. ernst hom-burg (Maastricht university), dr. Barend van der Meulen (university of twente and rathenau institute), prof. dr. arie rip (university of twente), prof. dr. Jan de Wit (radboud university nijmegen).
92
chapter 4
4.4 The changing societal contract of Dutch academic chemistry, 1975-2005
in the following we will empirically address our question about practical applications as a source of
credibility in two parts: a historical analysis of the changing science-society contract and a sociological
analysis of the position of practical applications in today’s credibility cycle. This section will describe
the way the content of the science-society contract has changed for chemistry. The main attention will
go to the conditions, as this part of the contract is most directly connected to the resources of research
practices. The conditions (as precipitated in the contract) concern the institutional environment of
academic research, which shapes the conversions of the credibility cycle. our analysis covers the period
1975-2005, but we will begin with a sketch of the situation in the 1950s and 1960s.
during the last decades the identity of academic chemistry has changed from basic research to the
production of strategic knowledge. although some professors already had strong ties with chemical
companies in the 1950s and 1960s (homburg & Palm 2004b), in these years direct contributions to
practical applications were not regarded as the main task of academic chemists. Since the introduction
of science policy in the 1970s and innovation policy in the 1980s, however, the government has increas-
ingly expected academic chemistry to address societal needs and to produce applicable knowledge as
well65. during the 1990s the idea of ‘strategic research’ has won ground, which concerns the develop-
ment of fundamental insights in domains of high relevance for economy or society (irvine & Martin
1984) The strategic identity of academic chemistry endures in the new millennium as it is compatible
with the most recent innovation concepts, in which the university is seen as a supplier of basic knowl-
edge which can be valorized by other actors in the innovation system.
in the rationales for funding academic chemistry the emphasis has shifted from education and
cultural value to the need for innovation and sustainability. in the first postwar decades the two domi-
nant rationales were the necessity of chemical research for the training of new r&d-workers and (less
importantly) the cultural value of basic research (homburg 2003, hutter 2004). in the 1970s the wake
of environmental awareness and the start of science policy together caused a shift in the rationales from
industry’s need of educated workers to society’s need of chemical expertise in the wider sense (rip &
Boeker 1975). The budget cuts on basic research in industry in the 1980s increased the importance of
the rationale for supporting academic chemistry related to its potential contribution to technological
innovations66 (de Wit et al. 2007, van helvoort 2005). in the 1990s the notion of sustainable develop-
65 in the first policy paper (M74), the primary mission of science policy is defined as enhancing the agreements of research agenda’s with societal demands.
66 The first foresight study of chemistry (vS80), commissioned by the minister of science policy concludes that academic chemistry should define its research goals more sharply and that the contacts with industry deserve intensification.
93
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
ment became increasingly significant in rationales for funding (chemical) research67. around the turn of
the century, the funding of university research was increasingly framed as support for the national inno-
vation system. in this perspective support for university researchers is accompanied with the expectation
that they actively interact with other actors in the innovation system, and contribute to the process of
“valorization”, by writing patents or by starting spin-off companies.
regarding the institutional conditions, during the 1950s and 1960s chemical scientists had a high
degree of autonomy. The most important types of funding (the so-called first and second money stream)
were distributed without any conditions attached, based on considerations of academic quality and
reputation. The first money stream was direct funding from the government to universities. The second
money stream was supplied by Son, the dutch research council for chemical research, founded in
1956, whose missions were to stimulate and coordinate basic chemical research68. Son’s resources were
distributed by ‘working communities’, thematic research networks, based on considerations of innova-
tiveness but with little steering power (hutter 2004). as a third stream of funding a small proportion of
all professors engaged in contract research for industry (homburg 2003).
in the 1970s scientists were increasingly held accountable for their work. although they were not
yet directly affected by policy measures, chemical researchers needed to put more effort in explaining to
society what they were doing. governmental science policy explicitly aimed at enhancing the agreement
of research agenda’s with societal demands (M74). initially this was attempted by simply facilitating the
interactions of scientists with societal actors, but in subsequent decades substantial shares of science’s
unconditional support were transformed into earmarked funding69.
from 1980 onwards a considerable change occurred in the funding of chemical science, shifting the
emphasis towards applicability (van der Meulen & rip 1998). This change was due to two intertwined
developments. first, the government was faced with the need for budget cuts on science, entering
into the era of ‘steady state’ funding (Ziman 1994). faced with the need for budget cuts on scientific
research, the chemical community proactively took several initiatives to advice about task division and
priorities. a committee of the chemical section of the dutch royal academy, installed in 1979, gave input
to the governmental generic report on task division and concentration in 1983. in 1983, science policy
minister deetman started the implementation of ‘conditional funding’, a system in which part of the first
money stream was subject to selection based on criteria of scientific quality and societal significance
(M84). The implementation of conditional funding was an occasion to both cut budgets and to increase
67 ‘for all new scientific and industrial activities on the field of chemistry the strives for sustainability and the mini-malisation of environmental pressure have become important boundary conditions.’ (original emphasis)(o95, p. 2). nWo’s Strategy note on chemistry for 2002-2005 is even titled: ‘chemistry, Sustainable and interwoven’ (cW01)
68 The aims of Son are ‘the enhancement of fundamental research at universities, colleges and other institutes in the area of chemistry in the broadest sense and the development of cooperation among researchers who carry out such scientific research.’ (hutter 2004).
69 By various policy instruments such as ‘priority fields’, national research Programs’ and ‘conditional funding’.
94
chapter 4
governmental steering of research directions. deetman explicitly mentioned that the connection be-
tween chemical research and societal needs should be improved (M83, M84). in contrast to other fields,
in the assessment of chemical research proposals, ‘social relevance’ was used as an important criterion
(Blume & Spaapen 1988). Moreover, informed by foresight studies (K82, vS80), an increasing share
of the second money stream was dedicated to application oriented research. Son started a program
for applied chemical research in 1980, together with the new technology foundation StW, the share of
which in Son’s total budget grew steadily to approximately 20% in 1995 (S95). in 1988 Son’s mother
organization ZWo was drastically reorganized into the new nWo, which resulted in increased funding
for application oriented research (van der Meulen & rip 1998, Kersten 1996).
Second, and most significantly, the third money stream, which was often application-oriented, grew
fivefold (see figure 4.2). during the 1980s chemical companies became more willing to sponsor aca-
demic research, due to their budget cuts on in-house basic research which made them more dependent
on basic research conducted elsewhere (de Wit et al. 2007, van helvoort 2005) (K84). also the minis-
try of economic affairs entered the scene and started the ioP-programs70 to fund university research, in
order to enhance the innovation capacity of the netherlands.
Figure 4.2 growing importance of third money stream in the funding of chemical research at dutch universities. numbers represent temporary paid fte’s in each money stream (ac91)
These developments continued in the 1990s. The second money stream continued to broaden its mis-
sion beyond basic research. The minister of science policy adopted the recommendation from the 1995
foresight study (o95) to strive for increasing the share of industrially steered chemical research from
50% to 75% (M97). The reorganization of nWo in 1998 implied the end of Son as a separate founda-
tion, which continued as the area chemical Sciences within nWo.
70 ‘innovatie gerichte onderzoeksprogramma’s’ (innovation oriented research programs), for example ‘Membranes’ (1983), ‘carbohydrates’ (1985) and ‘catalysis’ (1989).
0
200
400
600
800
1000
1200
1400
1600
1800
1980 1985 1990
�e
year
Temporary sta� third money stream Temporary sta� second money stream Temporary sta� �rst money stream
Permanent sta�
95
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
another significant change in the 1990s was the institutionalization of performance evaluations.
governmental policy makers, research councils and university managers who had all gained steering
power in the funding allocation, developed a need for transparency regarding research outcomes. in
1996 the first nation-wide quality assessment of chemical science was conducted (v96), the second in
2002 (v02). as an input to these evaluations, the research groups had to carry out a self-assessment,
dealing with the quality and impact of their work but also the societal value of their output. due to
the lack of guidelines for this issue, each group had its own style in presenting actual or potential ap-
plications of the knowledge it produced (van der Meulen & rip 2000). as a result, the self-reported
relevance seemed more a product of the group’s creativity than of its actual relationships with societal
stakeholders or contexts of application71. also in the final examination by the ‘visitatiecommissie’, the
evaluators were free to choose to what extent they take into account societal relevance or applications,
due to the lack of a strict protocol (van der Meulen & rip 2000, van der Meulen 2008). in practice, they
turned out to generally ignore this criterion and focus strongly on traditional scientific quality indica-
tors, like the number of publications in high quality scientific journals. relevance was one of the four
major criteria used (the others being quality, productivity and viability), but this was mainly conceived
as ‘scientific relevance’. The societal or economic impact of the research was only assessed if this suits the
group’s (self-defined) mission72.
after 2000 chemistry faced a further diversification of funding sources. Thanks to their contin-
ued growth, the european framework Programmes have become a substantial source of income for
academic chemists. Moreover, consortia-based funding emerged, large sums of governmental money
supplied to collaborative programs of university scientists which are monitored by (industrial) user com-
mittees. Significant examples are the nanoned program (2004) and the TTi ‘dutch Polymer institute’
(1997). Moreover, in 2002, the actS program was founded, for ‘advanced chemical technologies for
Sustainability’73. This program is funded by several ministries and chemical companies, but it is carried
out by nWo. today its volume is about half of all second money stream funding available for chemical
research in the netherlands: in 2005, the program amounts to 11.3 Million euros, compared with 14.5
Million for nWo-cW (cW06) and it anticipates further growth (a07).
71 e.g. the self-evaluation of chemical research in utrecht (u01). Some groups refer to specific societal contributions (like more effective drugs or cleaner industrial processes) and actual interactions with medical practitioners or industrial researchers; others claim to support a more applied scientific research field, such as biology, which in turn supports medical or pharmaceutical practice.
72 This may change in the near future, as there are attempts to develop more holistic evaluation methods that include societal relevance (Spaapen et al. 2007)
73 The program was initially defined narrower as ‘advanced catalytic technologies for Sustainability’, but soon the pro-gram was widened to a generic program for chemical technology.
96
chapter 4
Summarytable 4.2 provides an overview of the changes in the science-society contract of academic chemistry,
as has been discussed in this section. four important changes can be identified in the conditions under
which chemists work. first, the funding now available for university research provides room for con-
siderable efforts in application oriented domains, while in the first postwar decades, there was general
consensus that universities should restrict themselves to ‘pure science’. Second, the interactions with
industry have become more intensive. Because the main products of university research are not anymore
people (only), but also knowledge, industrial steering of the content of research has become justified.
Since the rise of innovation policy, industrial representatives have a say in the design of most major che-
mical research programs. Third, universities attempt to play an active role in the valorization of research
outcomes. Merely providing knowledge is not considered sufficient anymore. The national government
actively stimulates academic patenting and the creation of spin-off companies. although there is little
proof of actual success, all dutch universities provide facilities to support researchers in translating their
knowledge into commercial activities. fourth, systematic performance evaluations have become a po-
werful institution governing academic research. every research group is subject to regular assessments,
which tend to focus most strongly on bibliometric quality indicators.
Table 4.2 overview of the changing contract for academic chemistry (copy of table 3.1). + signs indicate that these ele-ments complement rather than replace existing elementsa.
Summary of identity Most dominant rationales Most important conditions
1950s and 1960s
Basic research educationcultural value
autonomyunconditional fundingSon communities
1970s + useful knowledge + Problem solving potential + Social accountability
1980s applicable knowledge technological innovation + conditional funding+ application oriented funding (StW, ioP, contract research)+ foresight+ Scarcity of resources reorganization nWo
1990s Strategic knowledge + Sustainable development further prioritization+ Performance assessments
2000+ Strategic knowledge + innovation system + consortia (actS, TTi, BSiK)+ european fPs
a actS: advanced chemical technologies for Sustainability; BSiK: Besluit Subsidies investeringen Kennisinfrastructuur Programs; ioP: innovation oriented Program; nWo: dutch organization for Scientific research; TTi: technological top institutes; Son: chemical research netherlands; StW: technology foundation
97
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
4.5 Academic research and practical applications: a credibility cycle analysis
how do the changes in the contract play out in the daily practice of academic researchers? have contributions to practical applications become rewarding in terms of credibility? in this section we will closely analyze the six major steps of the credibility cycle, with special attention for the differences across the three fields of chemistry we have investigated.
4.5.1 From recognition to money
Section 4 has shown that the palette of available funding sources has changed dramatically.
The three fields we have studied use a variety of funding sources (see table 4.3). do (promised) practi-
cal applications help them to acquire funding?
Table 4.3 overview of the most significant funding sources in the three sub-disciplinesa
Biochemistry catalysis environmental chemistry
funding sources nWoeu fPs
nWo, StWeu fPsindustryconsortiaentrepreneurship
nWo, StWeu fPsgovernmentindustryngos
a Based on our interviews.
in catalysis, promising a contribution to practical applications is a requirement for most types of fun-
ding. The procedures for acquiring funding vary. to get money from an individual firm, very short des-
criptions can suffice to convince of the quality and relevance of the proposed project. research councils
and hybrid consortia, however, demand extensive proposals addressing a number of predefined issues
like innovativeness, scientific relevance, societal impact, research methods, expected outcomes and
deliverables. except for nWo and the ‘national research School combination catalysis’, all funding
sources require that their research is both relevant for industry and excellent in scientific terms. Several
of our interviewees in catalysis have founded small firms based on patented inventions. currently these
are still too young to make profit, but in the future they may serve as sources of research funding. in case
such a company is acquired by a larger firm, a significant sum will flow to the research department, to be
spent freely on research activities. The increased industrial influence on the research agenda generates
incentives to pay closer attention to possible practical applications, but it does not imply a shift from
basic to applied research. Because catalysis is a technology to make chemical processes more efficient, it
98
chapter 4
carries the potential of making chemical industry more sustainable. Thanks to this promise and thanks
to the good reputation of dutch catalysis, both public and private funding sources are willing to support
research on fundamental questions:
‘And it is of course the case, that they seldom really let us do a research project in order to get that specific
catalyst after four year for sure. They are rather interested in having you work in a particular area of re-
search, of which we see together: this is promising. And then the innovations come automatically and if they
really want to apply it, they pick it up themselves.’ ( full professor, catalysis)
for environmental researchers, however, there are significant differences among the various possible
sources of funding. The national research council nWo and the european framework Programmes
(fPs), on the one hand, strongly focus on academic quality and reputation. governmental bodies, in-
dustry and ngos, on the other hand, have specific questions that can be answered by applied research.
These funders tend to look for the researcher who can answer them with the best price-quality ratio,
creating a competition between academic researchers and (semi-)commercial research institutes. The
projects for industry and other application oriented funding sources are often pretty short. They can
threaten the continuity in research activities, they involve little basic research and they are not suitable
for researchers to get a Phd-degree. however, environmental chemists need to do them to remain finan-
cially healthy, and they pay better than research councils74.
in biochemistry, scientists only use funding from sources oriented to basic research. The various
grants and programs of nWo are most significant; next to that the european fPs are gaining impor-
tance. acquiring money from industry seems hardly possible because ‘the time horizon of companies, also
of big companies, has become very limited’75. Academic quality of research proposals and of research groups are
still the most important criteria to get money from NWO and FPs. The rise of thematic priority programs does
not seem to have serious consequences. In many programs referring to potential practical applications enhances
the chances of success, but this can usually be dealt with by rather loose and unrestrictive statements. One
respondent explained that he simply looks around what thematic programs are available and then think up a
link with his own competences and existing research plans:
‘You try to make your expertise fit, we have become pretty good at that. You write it down in such a way
that it fits the program.’ ( full professor, biochemistry)76
74 researchers aim for a diverse range of funding sources, in order not to be dependent on one client (interview 11).75 full professor, biochemistry.76 This reasoning is in line with Morris’ observation that biologists tend to adapt their proposals to fit the priorities and
initiatives of funding bodies (Morris 2000, p. 433).
99
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
overall, the changes in the funding structures do not have strong implications for the research directions
biochemical scientists choose.
another way of turning practical applications into money is consultancy. in catalytic and in envi-
ronmental chemistry it is common to conduct small consultancy projects for public-sector or private
organizations beside the (bigger) research projects. consultancy is not common in biochemistry.
it must also be noted that the scores on official performance evaluations increasingly contribute to
the funding available to a research group. university managers take them into account when faced with
the need for budget cuts77. also in the review process of nWo proposals the scores are used as a quality
indicator. groups with a good score will advertise it when attracting contract research as well.
to conclude, in all fields an increasing share of all funding demands researchers to articulate possible
practical applications in industry or society of the proposed project. in catalysis and environmental
chemistry, however, the promises about practical applications tend to be much more explicit and specific
than in biochemistry.
4.5.2 From money to equipment and staff
once a scientist has acquired research funding, he or she can use it to buy equipment or to hire one or
more people to carry out the work. What criteria are used in the selection of candidates?78 do (realized
or promised) practical applications play a role? asked what characteristics they look at when selecting
candidates for academic positions, senior researchers mention research quality, abilities to attract fun-
ding and management and collaboration skills. Publication lists stand out as the most important quality
indicator. This is confirmed by all respondents, both juniors and seniors, for example:
‘When we hire new staff, their publication list is the most important criterion, possibly in combination with
a Hirsch-index or something similar. The same goes for contract extension and for promotion to associate or
full professor.’ ( full professor, environmental chemistry)
‘You’ve got to publish a lot. Imagine I have a post-doc position and there are 60 candidates. First I look at
the people who publish 10 articles a year. Why? Because they will also publish 10 articles for me.’ (associate
professor, catalysis)
Phd-students do not believe that practical applications of their research will enhance their prospects
for an academic career. however, in catalysis some professors report that they do look at the number
of patents a candidate has developed, or the amount of interest that industry shows for his research (in
77 This is the experience of most senior researchers we have interviewed.78 Purchasing equipment has not been investigated in this study.
100
chapter 4
terms of industrial funding). But practical applications are not crucial for every scientist. as one profes-
sor claims, it is simply important that someone is part of an international community:
‘(...) this international community can consist of pure scientists, but it can also consist of a mix of pure sci-
entists and people from industry. And this balance, for some it lies all the way to the fundamental scientists
and on the other hand it can also lie in the direction of applications. And I think both are just as valuable.’
(retired professor, catalysis)
another quality taken into account is the candidate’s proven acquisition skills:
‘Today it is important that you have acquired a European project once, or at least paid a significant
contribution to it. That you show that you can do that as well. But I would say that publications are really
number 1 and this is a good number 2.’ ( full professor, environmental chemistry)
for this reason, in catalysis researchers at various levels expect that contacts with industry will be valua-
ble in academic job applications.
to conclude, in the selection of candidates for academic positions, academic criteria rule. in cataly-
sis, career perspectives may be slightly enhanced by industrially relevant work or commercial activities,
but one’s patents are still far less helpful than one’s publications and citations.
4.5.3 From equipment and staff to data
What is the role of practical applications in the production of data?
in general, the collection of data takes place within the boundaries of projects that have been defined
in proposals or contracts. in practice, however, such descriptions function only as rough guidelines for
the actual research. on the laboratory floor, the boundaries between projects can be pretty fuzzy.
‘To the outside there is always a very clear line of demarcation. But within the group that demarcation is
much vaguer.’ ( full professor, catalysis)
not all projects offer the same degree of operational autonomy. a personal grant from nWo, for
example, is qualified as ‘reasonably free money’79, and is therefore highly appreciated. The same goes for
university funding. Money from industry or other third parties typically involve more communication
79 full professor, catalysis.
101
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
with the funding source, but the degree to which this decrease the flexibility for the researchers to devi-
ate from the original plans varies across fields.
in catalysis, firms providing (co-)funding obviously aspire to benefit from it, but they do not steer
the experimental work in detail. in principle, they do not predefine all details of the research to be
conducted, but only the type of system, class of compounds, or type of reactions to be studied. com-
panies hope to benefit from obtaining more background knowledge in the field they are working in,
which can serve as a source of inspiration for more applied innovation projects conducted in-house.
industrially (co-)funded projects usually have a supervising committee which receives an update about
the progress about three to four times a year and which can suggest particular directions, but these are
only followed if this does not hinder the academic development of the Phd-student involved. according
to our respondents few disagreements on this point occur. in cases in which companies steer a project in
specific directions this often has little implications for the academic question that is addressed. The same
catalytic mechanism can be studied using different substances.
The situation is different in environmental chemistry. here the projects for industry, the government
or ngos tend to be short and serve specific goals. in this domain specific actors often have a strong
stake in particular outcomes, which can complicate the collection of data80. researchers have an interest
in having the assent of all organizations involved (government, industry and interest groups), because
this increases the impact of their outcomes. however, this may challenge their independence, as some
parties may try to influence the outcomes to their own benefit.
Most biochemists we have interviewed do not have any contacts with possible users of their knowl-
edge. one professor regularly meets with medical researchers in the context of a research project with
medical relevance, but the others do not report any interaction outside their own field influencing their
work.
to conclude, during data collection biochemists are not concerned with practical applications at all,
but researchers in catalysis and environmental chemistry tend to interact frequently with industry or
other users that (co-)fund the research. in environmental chemistry such interactions sometimes disturb
the data collection; in catalysis this happens less, and they are often perceived as a source of inspiration
and motivation.
4.5.4 From data to arguments
although the conversion of data into arguments is relatively straightforward in chemistry, it is still an
active step with significant degrees of freedom. to what extent do practical applications influence this
80 an associate professor told us that in a project about the risks of a particular class of compounds his work was compli-cated, because of the sensitivity of the required information.
102
chapter 4
process? for academic chemists the main consideration in this step seems to contribute to scientific
debates. researchers use their data to construct claims that fit in a particular scientific discourse in which
they are participating. The arguments they develop are their tool for positioning themselves within a
particular research community (latour & Woolgar 1986).
in catalysis we found no evidence of the influence of practical applications on the arguments re-
searchers produce, apart from an emphasis on either environmental or economic benefits. in the other
two fields, however, the funding source of the research does influence the production of arguments. in
biochemistry arising practical applications can steer the arguments in a particular direction. The results
of biochemical experiments paid by a patient organization need not be of a different kind than the ones
from experiments funded by nWo, but the former are more likely to be converted into medical argu-
ments while the latter may be used only to contribute to more fundamental biochemical debates81. in the
case of contract research, the sponsor (or ‘client’) may also influence the (types of) arguments pro-
duced. The environmental chemists in our sample report that they sometimes have difficulties defending
their academic ‘objectivity’ against unwanted interference of the companies or environmental agencies
that have a stake in the research. industry often hopes the data are turned into good news about the
safety of chemical substances.
‘Yes, then it is difficult to act. Also when deriving conclusions.’ (associate professor, environmental chemis-
try)
But researchers try to keep their independence.
‘And maybe we should not be concerned too much with what the client wants.’( full professor, environmen-
tal chemistry)
to conclude, in most cases researchers are relatively autonomous in developing their data into argu-
ments, but in applied research projects the funding party sometimes succeeds to influence the conver-
sion of data to arguments, thereby harming the objective position of academic scientists.
4.5.5 From arguments to articles (and other outcomes)
Publishing in scientific journals is of vital importance in all three fields of chemistry. Many scientists
try to get their work published in the best journals possible, which is often defined as the one with the
highest impact factor.
81 This is visible in the research of a full professor in biochemistry.
103
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
‘Yes, it is important, for two reasons. First of course one wants to make ones findings publically known. This
is a way to receive recognition of your peers. Second it is also dire necessity, in order to secure the continuity
of funding. Because if you do not have publications... it is the way for the outside world to assess you.’ (as-
sociate professor, biochemistry)
‘I think it is the only way to show what you have done. And I think that if something is not publishable in a
scientific journal, it is not worth much.’ (PhD-student, biochemistry)
in principle, application oriented research can also be published. Both in environmental chemistry and
in catalysis, scientists publish the results of research issued by industry or other users in prestigious
journals, too. however, this is not always as easy as with research funded by the first or second money
stream. companies sponsoring catalytic research are protective with respect to commercially relevant
outcomes. research contracts usually specify a period in which a company has exclusive access to the
results to explore the feasibility of the developed technology and to consider applying for a patent before
the academics are allowed to make them public. This hardly ever leads to complete bans on publications,
but it does complicate early stage communication such as poster-presentations. There are also exceptio-
nal cases in which a research project is completely secret and no publications are allowed at all.
in environmental chemistry, the small size of many assigned projects complicates the publication of
academic papers.
‘Yes, then you almost always face the situation that it is just too little to make a good scientific publication
about it.’ (associate professor, environmental chemistry)
Moreover, writing a paper beside the project report requires extra work which can not be accommoda-
ted within the projects themselves.
in their evaluation of manuscripts, journal editors and reviewers hardly assess the (possible) practi-
cal applications of the research. in all three fields, the main criteria in the selection of papers are the
novelty, accuracy and the scientific relevance of the research. in environmental chemistry, it may help if
one manages to link one’s research to an important societal issue like global warming.
‘Climate change is of course very hot. So in the piece we are currently working on, we try to steer it a bit
in that direction. So that the result is useful. And sustainability. That you try to associate with the fashion
terms. [...] On the work itself it does not really have an influence. But it does on your introduction, how you
stage your story, sketch the framework, there you includes a couple of words.’ (PhD-student, environmental
chemistry)
104
chapter 4
in other fields, it is matter of personal style whether one refers to the societal context or the possible
applications of the research. Both in biochemistry and in catalysis some researchers make an effort to do
this, but this does not particularly help them to get their papers published.
although scientific publications are the most important type of research outcomes, chemists deliver
other products, too. in environmental chemistry it is common to write scientific reports for the organiza-
tions commissioning the research. catalytic chemists are frequently involved in patent applications.
Some senior researchers have contributed to tens of patents. Phd research commissioned by industry
often leads to patents, for which the companies sometimes even pay a bonus. in other cases academics
write patents themselves and start a company to make a profit of it, that partly be used as research fund-
ing. although less common, this phenomenon starts occurring in biochemistry, too82.
in conclusion, journal publications remain the most important form of output in academic chem-
istry, but under the new science-society contract researchers also produce patents and write scientific
reports. in environmental chemistry application oriented research is more difficult to publish due to the
small project size. in catalysis practical applications only create a delay, but do not inhibit the eventual
publishing of results. anyway, practical applications never help to get one’s work published as the selec-
tion of journal manuscripts is based solely on academic criteria.
4.5.6 From articles (and other outcomes) to recognition
do practical applications contribute to an academic reputation? one can distinguish a formal and an
informal component of recognition. The formal component is the result of official research group evalua-
tions and individual performance interviews. Someone’s informal recognition is based on the assessment
of colleagues of one’s qualities, expressed in conference contributions, (informal) discussions, and pu-
blications. Both types of recognition mutually influence each other. a good score on official evaluations
will be known by one’s colleagues and taken into account in their informal recognition of one’s work. in
turn, informal recognition also contributes to the score on formal performance evaluations.
Within one’s own small subfield one can earn recognition for the content of one’s papers and
lectures. These are valued for their innovative content and for being published in prestigious journals.
Beyond one’s own specialty one’s reputation is more based on quantitative indicators like publication
and citation scores and formal performance evaluations.
‘[...] in practice we measure impact-factors. And these we add up. Calculations are being done up to three
digits.’ ( full professor, biochemistry)
82 The interviewee involved (associate professor in biochemistry) claims that he was stimulated by nWo policy.
105
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
Since research grants have become scarce, the amount of funding attracted has also become a contribu-
ting factor to one’s reputation. in environmental chemistry especially the more academic grants, like the
ones from research council nWo are appreciated, as they are generally regarded more prestigious than
funds from the ‘third money stream’.
Practical applications of academic chemistry contribute little to recognition. Some catalytic re-
searchers have respect for contributions to industrial innovations, especially in engineering subfields.
however, these contributions often remain invisible to academic colleagues. The number of patents
scientists hold does not contribute much to their academic reputation; they can even have a negative
effect. one professor argued that his long list of patents tends to distract people from his academic suc-
cess and makes them forget that he also has an impressive list of scientific publications. in biochemistry
societal contributions do not play any role in getting academic recognition. Promises about applications
(most often in the medical domain) can contribute only indirectly to one’s reputation if they help to start
big research programs or consortia. Practical applications of environmental research also contribute little
to one’s academic reputation83.
Beside the issues discussed so far, some other aspects that may contribute to informal recognition
are management skills, collaborations with well-respected scientists, educational work, and presentation
skills. however, our interviews indicate that these are all of far less importance than journal publications.
With regard to formal recognition, all scientists have a performance interview with their direct boss
once a year. Practical applications receive very little attention84. research productivity is the most impor-
tant issue on the agenda. in environmental chemistry, also non-academic output, the so-called ‘grey’ pub-
lications, is taken into account. in catalysis and biochemistry these are not regarded as output. for senior
researchers contributions to education, funding acquisition and other management tasks are also discussed.
The scores on performance assessments both count for internal university policy and for the acquisi-
tion of additional funding in the second and third money stream. in the official protocol one out of the
four main criteria is ‘relevance’ (v03). in practice, however, the evaluation committee has the freedom
to choose its own interpretation of this criterion. in biochemistry, the committees typically define it as
scientific relevance, because this is considered most appropriate in a field of basic research. in catalysis
and environmental chemistry, however, the reports do not clearly define their concept of relevance and
do not express the extent to which it concerns societal relevance as well (v96, u01, v02).
83 two out of five respondents perceive them not to contribute at all. The three others claim that applications can make a contribution to one’s reputation, but this is complementary to one’s scientific impact. ‘everything with climate change of course is an example. if you find important new things there, you will receive many invitations to tell about it somewhere, both at scientific and at more societally organized conferences. But, on the other hand, you are invited just as often for scientific conferences if you simply have produced sound research and you have shown that you can give a nice talk about it.’ (associate professor, environmental chemistry).
84 one professor reports that his boss appreciates his publications in popular media and some of his additional functions because they contribute to the visibility and the impact of his institute.
106
chapter 4
to conclude, in none of the three fields practical applications significantly contribute to recognition.
recognition is mainly based on academic publications. Beside informal recognition, formal processes
like performance interviews and performance assessment also contribute to one’s reputation, but all
focus on the same quality indicators: publications and citations.
4.6. Conclusion: the importance of stakeholders
our analysis has shown that the relationship between academic chemistry and society has undergone
some major changes. under the current societal contract, academic chemists are expected to deliver
strategic knowledge and to participate actively in the valorization of research outcomes. due to the
changing demands of both public and from private funding sources, academic researchers have lost a
considerable degree of autonomy in choosing research agendas. at the same time, however, they are sub-
jected to practices of quality control which hardly reward applicable knowledge, spin-offs or patents, but
mainly publications in academic journals. in the performance assessments of 1996 and 2002 the societal
dimensions of academic chemistry receive little attention. although installed to increase the social ac-
countability of scientists, evaluations merely enhance the need for peer recognition. Bibliometric quality
indicators strengthen the pressure to publish in scientific journals and enhance the ‘publish or perish’
norm (Weingart 2005, Wouters 1997). as a result there is a potential contradiction between research
agendas that are fruitful with regard to funding acquisition and research agendas that promise to yield
peer recognition and high scores on formal evaluations.
have practical applications become a source of credibility? We found considerable differences
among the three fields under study. in catalysis, practical applications constitute a rich source of
credibility. Promising a contribution to industrial processes is a necessary requirement for acquiring
research funding. The intensive interactions with firms during the research process stimulate rather than
inhibit data collection and publications. Moreover, commercially viable outcomes can be turned into
new research funding by selling patents or exploiting them in a spin-off firm. in biochemistry, practical
applications do not help a lot in gaining credibility. The available funding sources provide incentives to
articulate possible practical applications, but this has a modest effect, as subtle cosmetic adaptations of
existing research plans usually suffice. due to the rise of bibliometric performance evaluations, biochem-
ists experience a much stronger pressure to publish than to contribute to practical applications. for en-
vironmental chemists, practical applications have a positive effect on some parts of the credibility cycle,
but a negative on other parts. The funding structure provides strong incentives for application-oriented
research and to contribute even more directly to practical solutions than before. however, relatively
short, application-oriented projects are not most fruitful in terms of scientific publications, evaluation
scores and academic recognition.
107
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
a partial explanation of the different effects of the changing contract on these three fields, can be
found in their socio-organizational characteristics. in environmental chemistry the task uncertainty
(Whitley 2000) is probably higher than in the other two fields. catalysis and biochemistry have
developed a convergent research agenda, in which there is considerable agreement about problem
definitions and theoretical goals. in the young field of environmental chemistry, however, there are less
standardized procedures and less certainty about intellectual priorities. The boundaries with neighbor-
ing fields of environmental science (toxicology, soil science, ecology) are weak. in spite of the small size
of environmental chemistry, its research activities are very diverse. in the terminology of Becher and
trowler (2001), environmental chemistry can be characterized as a ‘rural’ field, with a low people-to-
problems-ratio and with no sharply demarcated or delineated problems. catalysis and biochemistry are
more ‘urban’, in the sense that there are many researchers working on a narrow area of study, and there is
strong mutual competition for priority of discoveries. The weaker competition and higher task uncer-
tainty make it more difficult for environmental chemists to publish in prestigious journals. Moreover,
they make this field more sensitive to external steering of its research agenda.
however, the difference between biochemistry and catalysis can not easily be explained by their
social organization. Both fields have high mutual dependence and low task uncertainty (Whitley 2000),
they are both urban, and composed of ‘tightly knit’ communities (Becher & trowler 2001). The crucial
difference between the two seems to be the type of stakeholders they have outside university. catalysis,
on the one hand, has a strong relationship with a homogeneous set of ‘upstream end-users’ (lyall et al.
2004), namely chemical firms (see table 4.4). due to the high investments these firms make in their
industrial facilities, they have a long term perspective. They make enough economic profit to be able to
make substantial investments in relatively fundamental research. Moreover the relationship between
chemical industry and academic catalysis is characterized by high cognitive and social proximity (tijssen
& Korevaar 1997), which facilitates the knowledge transfer and alignment of research activities.
Biochemistry, on the other hand, hardly has any ‘upstream end-users’ (lyall et al. 2004). as its main
applications are in the medical domain, its main stakeholders are health care practitioners and patients.
But these both figure more as ‘downstream end-users’, as they do not directly interact with biochemical
researchers and they do not have any formal channels to influence academic research activities. Bio-
chemical researchers only interact with ‘collaborators’ and ‘intermediaries’ that represent the stakes of
these downstream users. These categories of stakeholders, however, provide little funding for academic
biochemistry, compared to the more fundamental research councils. The lack of upstream users explains
that practical applications do not form an important source of credibility for academic biochemists.
environmental chemistry has a heterogeneous set of upstream end-users. environmental policy
makers, firms and environmental ngos all have a stake in this research, and all provide a share of the
research funding. however, the time horizon of these users is relatively short. The knowledge needs of
firms and ngos with respect to environmental chemistry are usually related to short term problems,
108
chapter 4
dealing with the regulation of specific chemicals. Policy makers typically have a longer term perspec-
tive, as they invest in generic models for the regulation of different classes of chemical compounds. Still,
their time horizon is much shorter than that of the companies taking an interest in catalytic research.
although environmental policy itself could benefit from a perspective up to several decades, in practice
the time horizon of policy makers is often limited by election cycles.
Table 4.4 The different categories of end-users of each field
catalysis Biochemistry environmental chemistry
upstream end-users industry - Policy makersindustryngos
collaborators (other catalytic chemists)
Medical researchers other (more applied) environmental scientists
intermediaries research councils research councilsMedical charitiesPatient organizations
research councils
downstream end-users
industry health care practitionersPatients
Policy makersindustryngos
to conclude, in this paper we have explored the interplay between shifting funding sources and the rise
of performance evaluations in the case of dutch chemistry. The degree to which these developments
stimulate scientists to contribute to practical applications turned out to strongly differ across fields. in a
field with powerful upstream end-users who can afford investments in fundamental research, contributi-
ons to practical applications have become a source of credibility, but in fields lacking such stakeholders,
researchers can hardly earn credibility by involvement with practical applications.
109
Practical applications as a source of credibility: a comparison of three fields of dutch academic chemistry
Appendix A (chapter 4). Documents studied
abbreviation Publisher / authora Year title city
S73-S95 Son 1973-1995
Jaarverslag den haag
M74 Ministry of Science and education
1974 nota Wetenschapsbeleid den haag
M76-M97 Ministry of Science and education
1976-1997
Wetenschapsbudget den haag
ar79 academische raad 1979 Beleidsnota universitair onderzoek den haag
K80 Kncv 1980 tien researchdoelen
vS80 verkenningscom-missie Scheikunde
1980 chemie, nu en straks: een verkenning van het door de overheid gefinancierde chemisch onderzoek in nederland
den haag
K84 Kncv & vnci 1984 toekomstig chemisch onderzoek: een uitwerking van het rapport Wagner i voor de chemie
ac91 acc-evaluatiecom-missie
1991 evaluatie van de universitaire chemie in de jaren ‘80
amster-dam
S91, S93, S94, S96
Son 1991, 1993, 1994, 1996
Meerjarenplan den haag
K94 Kncv & vnci 1994 toekomstig chemisch onderzoek: universitair fundament voor industriële meerwaarde
o95 ocv 1995 chemie in Perspectief: een verkenning van vraag en aanbod in het chemisch onderzoek
amster-dam
v96 vSnu 1996 Quality assessment of research: che-mistry: past performances and future perspectives
M00 Ministry of educa-tion, culture and Science
2000 Wie oogsten wil, moet zaaien: Weten-schapsbudget 2000
den haag
n00 nvBMB 2000 verder met Biochemie en Moleculaire Biologie: Beleid voor een vitale Weten-schap,
nijmegen
u01 chemistry - utrecht university
2001 assessment of research quality utrecht
110
chapter 4
cW01 nWo-cW 2001 Strategienota 2002-2005: chemie, duur-zaam en verweven
den haag
v02 vSnu 2002 assessment of research Quality: chemis-try and chemical engeneering.
vi03 vnci 2003 vijfentachtig jaar vnci in vogelvlucht leidschen-dam
v03 vSnu, nWo and KnaW
2003 Standard evaluation Protocol 2003-2009 for Public research organisations.
cW03-cW05
nWo-cW 2003-2005
cW-Jaarverslag den haag
M04 Ministry of educa-tion, culture and Science
2004 focus op excellentie en meer waarde: Wetenschapsbudget 2004
den haag
n04 vno-ncW, vSnu & nfu
2004 Beschermde kennis is bruikbare kennis: innovation charter bedrijfsleven en ken-nisinstellingen.
v05 vSnu 2005 onderzoek van Waarde: activiteiten van universiteiten gericht op Kennisvalori-satie,
den haag
cW06 nWo-cW and actS
2006 chemie@nWo: naar een environment of excellence; Strategische koers 2007-2010
den haag
r06 regiegroep chemie 2006 Businessplan: Sleutelgebied chemie zorgt voor groei,
leidschen-dam
a07 actS 2007 actS Means Business: Second Phase actS Plan 2007-2011
den haag
M07 Ministry of educa-tion, culture and Science
2007 voortgangsrapportage Wetenschapsbeleid den haag
a acc: academische commissie voor de chemie (The committee for chemistry of the KnaW)actS: advanced chemical technologies for Sustainability (a research program)cW: chemische Wetenschappen (division for chemical sciences)KnaW: Koninklijke nederlandse akademie voor de Wetenschappen (royal dutch academy)Kncv: Koninklijke nederlandse chemische vereniging (royal dutch chemical association)ncBMB: nederlandse vereniging voor Biochemie en Moleculaire Biologie (dutch association for biochemistry and molecular biology)nfu: nederlandse federatie van universitair Medische centra (federation of dutch academic medical centres)nWo: nederlandse organisatie voor Wetenschappelijk onderzoek (dutch organization for scientific research)ocv: overlegcommissie verkenningen (committee for foresight studies)Son: Scheikundig onderzoek nederland (research council for chemistry in the netherlands)vnci: vereniging voor nederlandse chemische industrie (association of the dutch chemical industry)vno-ncW: verbond van nederlandse ondernemingen en het nederlands christelijk Werkgeversverbond (dutch em-ployers’ association)vSnu: vereniging voor Samenwerkende nederlandse universiteiten (association of dutch universities)
chapter 5.
113
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends85
Abstractin this chapter we describe and explain how the relationship between biology and society has changed in
the netherlands over the past 30 years, and how this has provided pressures and incentives for multi-
disciplinarity in two different fields. in paleo-ecology we observed a rise in multidisciplinary research
collaborations as a response to outside pressures to do more ‘relevant’ research, particularly in the area
of climate change. in toxicology we observed the opposite: multidisciplinarity is used to strengthen
the fundamental research on physiological mechanisms of toxicity. We conclude that multidisciplinary
research collaborations can serve various ends, ranging from enhancing the practical value of a funda-
mental field to strengthening the fundamental basis of an applied field.
5.1 Introduction
across the whole science system the importance of collaborations between researchers with different
disciplinary backgrounds is growing (hicks & Katz 1996, Morillo et al. 2003, hackett 2005). This deve-
lopment is regarded as one of the most profound changes currently affecting academic research systems
(gibbons et al. 1994 , see also chapter 2). The rise of multidisciplinarity is often considered beneficial
because it opens up new fundamental research lines, because it may help science to more adequately
address society’s knowledge demands and because it may enhance the economic impact of science.
for these reasons, increasing interactions between scientific disciplines are a key feature in influential
85 forthcoming as: hessels, laurens K., Stefan de Jong & harro van lente. 2010. ‘Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends’, in J.n. Parker, n. vermeulen & B. Penders (eds), col-laboration in the new life Sciences, ashgate, 37-62.
114
chapter 5
visions on the future of academic research like Mode 2 knowledge production, Post-normal Science
and the Third generation university (nowotny et al. 2001, funtowicz & ravetz 1993, Wissema 2009).
however, the factors supporting the rise of multidisciplinarity are poorly understood. This chapter aims
to alleviate this by a detailed analysis of case studies of two life sciences, namely toxicology and paleo-
ecology.
one can distinguish between factors which are internal to the disciplines involved, and factors which
arise externally to them. a well-known example of an internal factor that incites multidisciplinary collab-
oration is the incorporation of molecular approaches in the life sciences, which has enhanced the need
to combine expertises and skills from different disciplines (cassidy & radda 2005). next to internal
scientific dynamics, external changes in the societal context of academic research also stimulate multi-
disciplinary interactions. according to gibbons et al. (1994), for instance, both public and private actors
demand the production of more ‘relevant’ knowledge, which may require multidisciplinary approaches.
others point to policy instruments on the national and international level that directly stimulate inter-
and multidisciplinary research (lepori et al. 2007, Bruce et al. 2004).
in this chapter we focus on these external factors in particular. The question we address, thus, is how
multidisciplinary collaborations in academic biological research have been affected by societal changes.
We are not the first to raise such questions. however, earlier accounts of the influence of contextual dy-
namics on the rise of multidisciplinary collaborations are often of a generic nature, paying little attention
to the different ways in which specific scientific fields are affected (e.g. gibbons et al. 1994). The current
chapter works to overcome this limitation by comparing two biological fields in their specific societal
context. We empirically limited our analysis to the netherlands, and to the period between 1975 and
2005. We found interesting differences between these two fields within biology. This suggests that any
generalized claim about the changing relationship between biology and society is severely vulnerable.
in the following section (5.2) we present the theoretical framework that guides our analysis, introduce
multidisciplinary research collaborations and briefly review literature on the topic. a key point is the no-
tion of a contract between science and society to understand the changing societal context of the dutch
life sciences. Section 5.3 discusses the methods and case-selection. Section 5.4 presents our empirical
analysis of the changing ‘contract’ between biology and society; Section 5.5 addresses the rise of multi-
disciplinarity in toxicology and paleo-ecology. We conclude with a comparison of the outcomes of our
two case studies and a general reflection on societal influences on multidisciplinary research collabora-
tions (section 5.6).
115
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
5.2 Theoretical framework
a scientific collaboration can be defined as ‘the work of teams of scientists with shared goals, such as
formulating or testing particular empirical hypotheses, and with shared products, such as co-authored
papers’ (griesemer & gerson 1993). Multidisciplinary research collaborations involve individuals
representing different scientific disciplines. Multi disciplinarity belongs to the same family of concepts as
inter- and trans-disciplinarity. in multidisciplinary collaborations, each discipline remains autonomous,
while the participating disciplines each contribute to the solution of a common problem. The collabo-
ration does not change existing disciplinary structures ( Jantsch 1972). interdisciplinarity implies closer
interaction than multidisciplinarity. interdisciplinary collaborators explicitly formulate a ‘discipline-
transcending terminology or a common methodology’ (gibbons et al. 1994). transdisciplinary research
goes even further; it requires a common theoretical understanding and is accompanied by interpenetra-
tion of disciplinary epistemologies (lawrence & despres 2004).
This definition of multidisciplinarity calls for clarification in two directions. first, it makes multi-
disciplinarity a relative concept, depending on one’s definition of a discipline. in this paper we speak
of biology (or the life sciences) as a discipline and of paleo-ecology and toxicology as fields. We regard
interactions within fields, which may include various specialties, as regular collaborations, but speak of
interactions among different fields or different disciplines as multidisciplinary collaborations. a second
aspect that requires clarification is how scientists relate to their discipline or field. in this chapter, one’s
disciplinary background refers to one’s current work rather than to one’s education. education strongly
influences the direction of one’s research, but still researchers can switch and develop careers in other
fields than those in which they were trained. an indicator to consider for a researcher’s field could be his
or her current affiliation. however, this can be deceptive as well. Bourke and Butler (Bourke & Butler
1998) have shown that a significant proportion of publications within a discipline originate from authors
affiliated to institutes of a different disciplinary classification. Therefore, we consider a researcher’s own
perception of the field to which he or she belongs as the most reliable indicator.
although interactions between scientific disciplines have occurred for centuries,86 the relative occur-
rence of multidisciplinary research collaborations has recently grown in all fields of science. The number
of journals that cannot be assigned to a single scientific field grows faster than the number of single-field
journals (hicks & Katz 1996). Moreover, since 1980 the number of papers that use the term ‘interdis-
ciplinarity’ or ‘multidisciplinarity’87 has grown exponentially (Braun & Schubert 2003). finally, papers
increasingly cite publications in journals outside their own field (Porter & rafols 2009).
86 in a sense, biologists were already involved in multidisciplinary collaborations in the 17th century when they joined expeditions into the unknown world in order to collect new species (Magner 1994).
87 The set of these papers contains both multi- and interdisciplinary studies and studies about the (desired) rise of multi-disciplinarity and interdisciplinarity.
116
chapter 5
The appeal of multidisciplinary collaborations has increased against the background of the increas-
ing specialization of sciences in the 1950s and 1960s (Klein 1990). in the life sciences pleas for multidis-
ciplinarity are often based on the conviction that the most pressing scientific and societal questions can
only be solved by combining knowledge and techniques from different disciplines. The human genome
initiative, for example, was set up as a multidisciplinary research effort, bringing together technologies
from biology, computing, material science, instrumentation, robotics, physics and chemistry, in order to
create tools for the analysis and understanding of human genes (tinoco 1987). The benefits expected
of multidisciplinary collaborations in the life sciences range from improved management of fishery
resources (Thakur et al. 2008) to the improvement of human health (Penders et al. 2009).
Previous research has provided insight into why scientists choose to collaborate, and why collabora-
tions are of growing importance. The most important reasons for scientists to collaborate are access to
expertise, improving access to funds, obtaining prestige or visibility, learning tacit knowledge about a
technique, pooling knowledge for tackling large and complex problems, enhancing productivity, mentor-
ing junior researchers and fun or pleasure (Katz & Martin 1997, Melin 2000, Bozeman & corley 2004).
The growing frequency of collaborations over the past few decades is also explained in part by increas-
ingly complex and costly instrumentation, decreasing costs of travel and communication, the impor-
tance of social interaction for the cognitive advance, increasing specialization, the growing importance of
interdisciplinary fields and political factors such as european integration (Katz & Martin 1997, laudel
2001). More recent studies also point to the stimulating influence of funding arrangements on collabora-
tions (vermeulen 2009, Shrum et al. 2007).
in general one could argue that the rise of multidisciplinary biology may be caused by both internal
and external factors. internal factors are social, cognitive or technical developments within biology. to
illustrate, the development of new research systems, such as molecular visualization techniques, created
the need to bring together expertise of different scientific disciplines. Biology and medicine provide the
research questions, but physicists and engineers develop the hardware, mathematicians and computing
scientists deliver software and analytical tools, and chemists and biologist develop contrast agents (cas-
sidy & radda 2005, loging et al. 2007).
Besides factors in science itself, societal developments can also stimulate multidisciplinary collabora-
tions in a number of ways. first, research councils and other funding sources nowadays exert a greater
pressure on academic science to deliver ‘relevant’ knowledge, knowledge that is practically useful, either
in the public or private sphere (see chapter 3). for instance, the rise of biotechnology has created a new
context in which the outcomes of academic research can be turned into commercial applications, and
urgent social problems such as climate change require new insights from the life sciences. in these cases
traditional disciplinary research may not suffice. Second, governmental science and innovation policy
may directly aim to directly stimulate multi- and interdisciplinary research. Since the 1970s, research
117
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
councils have developed funding instruments to stimulate multidisciplinary collaborations,88 both to
address complex societal issues and enhance fundamental understanding. Moreover, the european
framework Programmes supply funding for collaborative projects and programs that often include
multiple disciplines (Bruce et al. 2004). Third, the rise of the World Wide Web and related communica-
tion technologies facilitate more diverse and distributed forms of collaboration (vasileiadou 2009). The
center for the development of a virtual tumor, for example, successfully brings together researchers
from all over the globe to collaborate on computational and mathematical cancer modelling (Sagotsky et
al. 2008). These external factors are the focus of this chapter.
to examine the changing societal context under which the academic life sciences operates and its
influence on multidisciplinary collaborations, we employ the sociological notion of a ‘contract’ between
science and society (Martin 2003, elzinga 1997, guston 2000 , see also chapter 3). This contract is not
a physical entity, but a representation of the moral positions that encompasses all implicit and explicit
agreements between academic science and governmental departments, ngos, firms and other societal
parties,89 specifying what science should do (identity), why it should do this (rationale), and the appro-
priate conditions for science to function well (conditions), see figure 5.1.
Figure 5.1 The contract between science and society (adapted from figure 3.1)
according to this contractual perspective, the identity of science is connected to the provision of a valu-
able public good: relevant research outcomes. Science’s task is to produce knowledge and to deliver it in
forms like papers, patents, artefacts or educated people. The precise type of knowledge (basic, applied)
which is expected and the method of deliverance specified by the contract vary over time and across
disciplines. The contract, that is, the set of implicit and explicit agreements, also describes why science
deserves support. academic research is often regarded as a necessary condition for sustaining a system of
higher education, commercial product development and informing complex decisions and innovation.
The third element of the contract contains agreements about the conditions under which scientists work,
including expectations regarding the social structure of the research community, allocation of research
funds and incentives for producing more socially relevant knowledge.
88 e.g. The dutch research council nWo (see report n97 in appendix a).89 The concept of the science-society contract is further elaborated in (hessels et al. 2009).
Identity
Conditions Rationales
118
chapter 5
although one may speak about a contract between ‘science’ and society in general, the notion is also
suitable for analysis of specific scientific disciplines, as each discipline promises different types of applica-
tions and is linked to different societal stakeholders. Moreover, some disciplines are more application
oriented than others, or interact more closely with stakeholders. applied disciplines (like biotechnology
and medicine) may have a stronger multidisciplinary nature than basic disciplines (such as mathematics
and physics) because solving practical problems requires knowledge from several fields. however, cutting
edge fundamental research also often occurs at the borders of two or more fields, so multidisciplinary
collaborations may be just as important in disciplines working mainly to produce basic knowledge.
5.3 Research Approach and Methods
Case selectionin this chapter we analyse the external factors that incite multidisciplinarity in biology. We take the
idea of a contract between science and society, or between science and a discipline, as a starting point
and raise the question whether and how a change in contract will affect fields of biology differently. to
maximize comparative explanatory power, we have selected two biological fields that differ substantially
in terms of the orientation of their research agenda. traditionally paleo-ecology has a more fundamental
research agenda than toxicology. in the quadrant model of Stokes (1997), toxicology traditionally fits in
the ‘edison Quadrant’, while paleo-ecology fits in Bohr’s quadrant (see figure 5.2).
Figure 5.2 Positioning of paleo-ecology and toxicology in relation to Pasteur’s quadrant (based on Stokes (1997))
toxicology is a typical example of a field that is inspired by considerations of use. it studies the (adverse)
effects of compounds on organisms. it contributes to the risk assessment of chemical substances, and
also informs medicine and pharmacy about side effects of drugs (hayes 2007). traditionally, toxicologi-
cal research questions are related to practical issues and are not strongly inspired by a quest for funda-
Bohr Pasteur
Edison
Que
st fo
r fun
dam
enta
l und
erst
andi
ng
Considerations of use
Paleo -ecology
Toxicology
119
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
mental understanding, but within toxicology there is increasing interest in the study of the underlying
mechanisms of toxicity.
on the other hand, paleo-ecology, that is, the study of ancient ecosystems and environments, is
inspired by a quest for fundamental understanding. in contrast to toxicology, paleo-ecology has histori-
cally contributed little in the way of practical applications, instead providing fundamental insights into
biological evolution and taxonomy. although our analysis in this paper will show that this has recently
changed, considerations of use traditionally did not strongly influence the paleo-ecological research
agenda.
Methodsdata for this study are drawn from in-depth interviews and documentary analysis. We conducted 18
semi-structured interviews, each lasting about one hour. respondents included 15 academic resear-
chers of various ranks, working at three different dutch universities (table 5.1). Questions were asked
about their (collaborative) research activities, personal motivations, funding sources, and performance
evaluations.
Table 5.1 distribution of respondents over fields, universities and academic ranks
Paleo-ecology (8)toxicology (7)
utrecht university (10)university of amsterdam (3)Wageningen university (2)
full professor (4)retired extraordinary professor (1)associate professor (3)assistant professor (2) Post-doc researcher (1)Phd-student (4)
in order to triangulate and complement our findings in documents about the allocation of public fun-
ding and their impact on multidisciplinary collaborations we have also spoken with key-representatives
of three prominent biological research councils in the netherlands: Bion, nWo-alW and ZonMw.
Bion (dutch foundation for Biological research) was founded in 1970. it has funded biological
research until 1994, when it was reorganized into the foundation for life Sciences (SlW). in 1998 SlW
became part of the department of earth and life Sciences (alW) of the new dutch organization for
Scientific research (nWo). ZonMw is the dutch research council for medical and health research and
exists since 2001. all interviews were recorded and transcribed. The interviews with academic resear-
chers were analyzed using nvivo 8, a software package for qualitative analysis. a previous draft of this
paper has been distributed among all interviewees in order to validate our findings.
appendix B lists the 82 documents studied. The main purpose of these documents is to provide
insights in the changing contract between biology and society. documents were selected in order to
represent the positions of the most important parties that take part in this ‘contract’. academic research-
ers are represented by the royal academy (KnaW) and by various foresight committees; the govern-
120
chapter 5
ment by policy reports of the ministry of science and education and the ministry of economic affairs.
furthermore, intermediary organizations like research councils and the association of dutch universities
(vSnu) are represented by annual reports and other documents. The documents were screened for
statements about the three elements of the contract: the identity of biology, rationales for supporting it
and the conditions under which it should function.90 Besides, a sub-set of documents provided factual
data about the institutional conditions of academic biology, mainly concerning shifts in research funding
and the development of performance evaluations. Based on the collection of these statements and data
a history of the biology-society contract was written. our documentary analysis begins in 1975 because
this year roughly marks the birth of dutch governmental science policy (W7491).
5.4 A changing contract between biology and society
Science does not proceed in isolation. in contrast, it is embedded in a wider set of institutions, privileges
and expectations. The fundamental ideas about what science is (identity), why it should be supported
(rationales) and how this should be arranged (conditions) are the three fundamental ingredients of what
we have labelled as the contract between science and society. The contract is historically situated and
will be different for different disciplines. in this section we will trace the major changes of the contract
between biology and society since the 1970s, when science policy was introduced in many western
countries. in our analysis we will follow the three lines of the contract: the identity of academic biology,
the rationales to support it in the first place and the conditions to do so. in the next section we will
inquire how paleo-ecology and toxicology have responded to these general changes.
Identity: from basic to strategicin reports and documents of the 1970s academic biology is mainly positioned as a field of basic research,
not directly connected to any practical problems or needs. The first research assessment of biology
(r80) stated that most research can be qualified as purely scientific, not explicitly designed to lead to ap-
plication. in this period research council Bion (Stichting Biologisch onderzoek nederland) only fun-
ded ‘purely-scientific’ research which addressed questions of ‘purely fundamental nature, not inspired by
any immediate societal issue’ (B75: 3). during the 1980s, however, biology, like other sciences, became
increasingly defined in relation to its contribution to society and the economy. The relationship with
practical applications became stronger and especially the commercial success of biotechnology firms
underpinned the general promise of the life sciences. in the discourse of the policy documents, strategic
plans and evaluations biology has shifted from a ‘basic’ discipline to a project with huge potential
90 This analysis was conducted manually, without the use of a formal coding scheme.91 in the following we will use abbreviations such as W74 to refer to the documents listed in appendix a.
121
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
economic impact. in 1980 Bion started to fund applied research for the first time.92 Moreover, Bion
distributed an increasing share of its money in programs dedicated to five ‘priority themes’ that relate
to a societal need or to expected economic profit: molecular cell biology and genetics, biotechnology,
gerontology, soil biology and toxicology (B86). foresight studies and evaluation reports of the 1980s
and 1990s increasingly expressed the value of biology in terms of its contribution to society and the
economy. The 1985 foresight study by the dutch royal academy of the Sciences argued: ‘also because
of biology, current prosperity and current wellbeing of humankind has been achieved and will be main-
tained’ (r85: 211). and a performance evaluation of the life sciences in 1994 stated:
‘Biology, in addition to its own scientific raison d’être, is a basic, - if not the basic - discipline underly-
ing medicine, agricultural production and environmental sciences, all areas of knowledge essential to the
survival of mankind. Therefore, biology is likely to become the central discipline among the natural science
in the next century’ (V9: preface).
By now, it has almost become a cliche to coin the 21st century as the ‘century of biology’. This change in
identity has affected the way biology is stimulated and assessed. at the end of the 1990s, for instance,
‘collaborations with industry’ were introduced as an indicator for the strength of biological research in
assessment reports (v99).
a second shift in the shared understanding what ‘biology’ is and does, relates to the rise of multidis-
ciplinary linkages. in the 1970s biology consisted of a collection of semi-autonomous units. The dutch
research council for biology Bion, for instance, was divided into sub-disciplinary ‘Working com-
munities’, such as functional Morphology or neurobiology. interactions occurred, of course, but were
not seen as valuable per se. over the years, however, linkages between fields were increasingly seen as a
goal in itself, as something to be cherished and supported. during the 1980s Bion made agreements
to foster such linkages with the research councils for agricultural science and environmental science.
in 1988, it even reorganized its divisions in a multidisciplinary way (B90). The increasing interest in
multidisciplinary life sciences was broader than Bion only. in the 1990s the dutch royal academy of
the Sciences (KnaW) noted in biology a growing exchange of knowledge between the different levels of
aggregation – from cell to ecosystem (vB97). The 1999 research assessment committee stressed that it
‘was very pleased to notice a growing integration of the knowledge acquired in different areas of biology’
(v99). The new research council nWo regarded the rise of multidisciplinary research as an important
development of the preceding decades (n95). around the turn of the millennium nWo’s department
for earth and life Sciences reported a growing willingness of others to cooperate, including nWo part-
ners and external partners who believe in the application potential of the produced knowledge (a98). a
92 in cooperation with the new technology foundation (StW) (B80)
122
chapter 5
KnaW foresight committee even considered biology ‘the nourishing core of a new kind of multidisci-
plinary research on living matter’ (c0: 11), offering inspiration to other sciences.
to conclude, in the past 30 years, the identity of academic biology has undergone two major
changes: first, it has lost it appeal of basic inquiry and has become connected with economic benefits;
second, biological research is increasingly seen as multidisciplinary research.
Rationales: from education to applicationThe element ‘rationale’ in the contract between society and biology refers to the question why society
should fund academic biology in the first place. traditionally, the (implicit) answer to this question was:
education. across the whole period we have studied, both researchers and policy makers have argued
that cutting-edge biological research is required to sustain high quality education in biology, both at high
school and higher levels (e.g. r83, W84, v99, c01). gradually, we noticed a shift to another rationale:
potential practical applications. The contributions that the life sciences can make to biotechnology, agri-
culture and public policy are increasingly used to justify investments in academic biology. already in the
1980s, such justifications were used when biologists claimed that their knowledge would lead to impro-
vement of food supply and health care by providing industry with a broad range of improved enzymes
(B85). furthermore, it was argued that biology provided other sciences, like biotechnology, agricultural
science and medicine, with fundamental knowledge (rip & nederhof 1985). The 1985 foresight study
by KnaW, written against the backdrop of imminent budget cuts, used this argument to emphasize the
importance of biology for human welfare:
The quality of human life is at stake in medical research, while the quality of the environment can be
improved by restoring the disturbed balance in nature using the results of environmental research. In this
way biological research, the foundation supporting all this research, is a source of wellbeing and prosperity
for human life and for the appreciation and the conservation of the quality of life of plant and animal on
planet earth. (R85: 209)
The 2001 foresight study explicitly referred to the societal impact of biology as an argument for incre-
asing governmental funding: ‘Biology as a core discipline and the growing position of the biological
sciences in society are not yet acknowledged enough in terms of financial means’ (c01: 41).
Conditions: from mono- to multidisciplinary fundingas in other scientific fields, the circumstances under which biologists work have changed significantly
over the past few decades. The mosaic of available funding has shifted, creating a stronger impetus for
research with practical ‘relevance’, as discussed above. in addition, the funding allocation by research
councils (Bion and nWo) has changed from a rigid structure of monodisciplinary Working com-
123
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
munities to a flexible organization dominated by multidisciplinary research programs. Since the early
1970s research councils have stimulated multidisciplinary collaborations, in principle. interdisciplinary
cooperation between biology and biochemistry was one of Bion’s aims when founded in 1970. The
realization of collaboration among different disciplinary Working groups, Project groups and Working
communities both within and outside Bion has always been part of its explicit mission (B75). howe-
ver, Bion’s money was initially allocated by disciplinary ‘Working communities’, with few conditions
attached and little formal oversight (Kersten 1996). later Bion used instruments like the ‘dwarsver-
bandcommissie’ (cross-cutting committee) and the ‘priority areas’ in order to stimulate promising young
interdisciplinary fields with the potential to develop into independent research areas. Yet, such measures
did not break down disciplinary boundaries as individual researchers and groups tended to defend
their own disciplinary interests.93 in 1992, however, nWo, which had succeeded Bion, announced
that one of its long-term aims was building connections between biology and other sciences (e.g. the
environmental sciences and behavioral sciences; n92). Since 1994 research programs, not restricted to
disciplinary Working communities, have become the dominant form of organising research funding.94
The quality of proposals was no longer assessed in terms of individual projects but in terms of multidisci-
plinary programs in which the individual projects could fit.
The second shift in conditions is the establishment of an elaborate system of performance evalua-
tions. in the dutch university system research quality assessments have developed in the past 25 years
from informal deliberations to formal procedures with a fixed protocol (van der Meulen 2008). after
an ‘open deliberation’ conducted in 1978 (r80) and a foresight study in 1982 (r83), the association of
dutch universities (vSnu) organized systematic research assessments in 1992 and 1998 (v94, v99).
in the course of years, increasingly specific criteria and indicators have been developed to guide the pro-
cess. The most powerful indicators are by far the bibliometric indicators, such as the number of articles,
their citations and the impact factors of the journals. The later evaluations also took into account the
societal relevance and impact of the research but these aspects did only partly account for one of the four
criteria. The main effect of the rise of quality assessment is an increased pressure to publish in scientific
journals.
The third relevant change is the foundation of research schools at dutch universities, which also
contributed to the rise of multidisciplinary research in biology. it became important for researchers to be
member of such schools, and these favored multidisciplinary research connections. a major task of the
research schools is to provide training for Phd students and in the past decade Phd students in biology
have been educated in the context of a broad research theme rather than a specific scientific field (c01).
93 interview with former adjunct-director of Bion94 This change was implemented when Bion was reorganized into the foundation for life Sciences (S94).
124
chapter 5
5.5 Two biological fields under pressure
While biology in general has faced some substantial changes, it remains an empirical question how this
affects different biological fields. are the responses to the urge for practical relevance the same? how do
fields relate to institutional changes like shifts in available funding and the rise of performance evalua-
tions? do they benefit equally from the general promise of biology? in this section we will discuss the
changes in the field of paleo-ecology and toxicology. in Section 5.6 we will compare and contrast the
findings and relate them to our notion of contract and the analysis of multidisciplinarity.
Paleo-ecologyPaleo-ecology has traditionally had a descriptive orientation, aiming to enhance the fundamental
understanding of ancient ecosystems and to contribute to biological taxonomy. it tends to be inspired by
curiosity, or in the words of a retired professor, by ‘fascination’:
To find out what these plants looked like in the past, how evolution has been, what kinds of processes have
taken place. That is fascinating. I always liked it. It remains fascinating. (retired professor)
The traditional image of paleo-ecologists is one of patient researchers spending many hours looking
through their microscopes. compared to other life sciences, the work is relatively individual. research
collaborations within paleo-ecology were rare, and multidisciplinary collaborations even more so. The
one area of applied research to which the field contributed was locating oil fields for the fossil fuel in-
dustry. This research was traditionally located outside the universities, but against the background of im-
minent budget cuts and reorganizations,95 academic paleo-ecology started to conduct research projects
for oil companies, which could serve as an additional funding source.
We simply saw good opportunities in it. Also because our field was a bit too... because it has always been a
field in the twilight between geo and bio and always threatened to become the victim of the next reorganiza-
tion, because we are simply standing in the periphery. So then it also was a means to keep people active here
at university and also to maintain a bond. ( full professor)
The decreasing amount of university funding was an important driver for the move towards industry.
aiding fossil fuel companies did not require intensive collaborations with other disciplines. in utrecht
95 in the early 1980s the ministry of science and education implemented a substantial austerity policy (the policy of ‘task reallocation and concentration’, announced in W82) which has resulted in a 15% budget cut for universities, amount-ing to 285 Mfl. (about 128 Million euros).
125
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
the contract research was organized in a foundation that was closely linked to the research laboratory for
paleo-ecology, which even carried the same name.96
Significant shifts in the available funding have gradually decreased the possibilities for paleo-ecol-
ogists for conducting basic research. apart from the decline of direct support from universities, the re-
search council, as we discussed above, has also changed its policy and dedicated an increasing share of its
funding to projects and programs that are ‘application oriented’ or in areas of ‘strategic interest’. against
this background, a new area of potential applications emerged for paleo-ecology in the 1990s. Support
for paleo-ecology was increasingly framed in relation to climate change. With the succession of different
generations of environmental policy, the perspective has shifted to a global scale (grin et al. 2003). in
the 1970s and 1980s the emphasis was on mitigating effects of local environmental problems such as
surface water pollution (e.g. W76, W87), but during the 1990s the emphasis shifted towards preven-
tion, and a more global perspective emerged, with issues like global warming and biodiversity (W96).
after the rio de Janeiro treaty of 1992, there were increasing calls for knowledge about biodiversity.
The topic of ‘global change’, including environmental problems, nature conservation and biodiversity,
is often listed as topic of high policy relevance (c01). The government offered funding for optimising
the conservation of collections and biodiversity data, strengthening taxonomic research and setting up
a national herbarium (M95). Paleo-ecology received support for reconstructing paleo environments,
which could be used in long term climate and biodiversity forecasting (W95, a00). The new field of ap-
plications added a sense of change to the research work:
The basis is still the same, this remains first. But then you will do something with it. And I think that that is
a fundamental change. (retired professor)
So, thanks to the issue of climate change, the potential contribution to policy making remains an im-
portant rationale for funding paleo-ecology. in two strategic reports (n01, n06) nWo mentions some
paleo-ecological topics as important themes in life sciences: global change, biodiversity and the connec-
tion between the biosphere and the geosphere and the extinction of species. These topics are part of the
nWo theme ‘System earth’. a recent foresight study claims that biogeology, of which paleo-ecology is a
part, is of growing interest of policymakers (g03). This is also stressed by one of our interviewees:
In the 80s, it was generally believed that everything was known about fossils, so funding was hard to get.
Nowadays, since there is renewed interest in fossils because of the climate change debate, it appeared there
are big gaps in collections and it is recognized that these fossils are essential for recognition and analysis of
fossil materials. (post-doc researcher)
96 The foundation was simply called ‘laboratory for Paleo-botany and Palynology’.
126
chapter 5
The new appeal of paleo-ecology has enhanced the need for multidisciplinary collaborations: paleo-
ecological research can make an important contribution to the understanding of climate change, but it
cannot do so in isolation. cooperation is required with other life sciences like microbiology and ecology,
but also with geo-sciences.
We are not directly application oriented. So what we often do is, we contact people of whom we think: these
could apply this. For example, climate results related to climate change, there we actually contact, as soon as
we know how it is, climate modellers and they can then use it for the models and even test whether it is right
as well. ( full professor)
The sense of contribution paved the way for paleo-ecologists to participate in unexpected academic
endeavours, such as efforts to develop renewable energy solutions. an example is a multidisciplinary
project studying the possibility to grow water ferns as a source of bio-energy:
We need an engineer who can calculate something like this, like what are the technical details. You have to
dry this, you have to burn that, you have to do this and that. Of course that is not our core business. We
can only say: look, in the past we already had the situation in which we say: a lot of carbon is stored, is it
possible that we say this plant produces that much biomass, that we can obtain energy from it again? ( full
professor)
The new status of paleo-ecology culminated in 2004 in the foundation of the darwin centre, a major
research facility in biogeology (g03, a04). a KnaW foresight committee had argued that knowledge
about biogeochemical processes is essential for understanding human-induced changes in System earth.
The darwin centre was founded with the explicit intentions to provide practically useful insights.
Beside its academic mission, the centre should act as a knowledge consultant to government authorities
(g03). it appears to stimulate multidisciplinary paleo-ecological research by funding high-risk research
and new combinations of research. according to the researchers involved, this has resulted in more
multidisciplinary work and multidisciplinary staff. Multidisciplinary work is used to answer research
questions from multiple perspectives, to do research more efficiently and to validate data with comple-
menting techniques and methods: to increase reliability of studies. Within the centre, Phd-students
have multiple supervisors from various fields. The students regularly discuss their work in interdiscipli-
nary meetings which stimulate them to switch between different disciplinary perspectives.97 although
the centre is not embodied in a physical building, it does make it easier to frequently interact:
97 interview Phd-student in paleo-ecology: ‘one is sitting there with a team of like... 10 people for sure... discussing often. “What have you done, lately? What can we improve upon it?” of course we have a number of chemists in there and they say: “so, maybe you can use this chemical compound to do a particular method”.’
127
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
Now I can also walk down here easily to organic geochemistry, what is obviously something very different
from what we are doing here. ... Because yes, of course one gets results from something one is not an expert
in oneself, and then it is very nice if one gets feedback from those who do know about it, I must say. (PhD-
student)
to conclude, in the field of paleo-ecology multidisciplinary collaborations have increased due to its
relevance for a new, global agenda on climate change. to ensure sufficient income, academic paleo-
ecology cannot rely on fundamental research agendas only. in the 1980s application oriented research
for the fossil fuel industry provided extra income, but did not require multidisciplinary collaborations.
The contribution to the research agenda of climate change, however, requires intensive cooperation with
other fields.
Toxicologyin contrast to paleo-ecology, toxicology has had an applied orientation since its emergence, which
is connected with the rise of environmental concerns in the 1970s. in its early years toxicology was
strongly related to practical applications, such as risk assessment and risk management. incidents with
industrial toxic compounds and concerns about environmental problems in the late 1970s and the 1980s
raised questions about the risks of synthetic chemicals. academic toxicological research was expected
to directly contribute to the development of environmental policies for regulating hazardous chemicals.
The practical orientation is reflected in definitions of the field. in an influential handbook, for example,
toxicology was defined as: ‘the study of the effects of toxic substances occurring in both natural and
manmade environments. The main task of environmental toxicologists is to assess objectively the risk
resulting from the presence of such substances’ (duffus 1980), cited by halffman (2003).
however, the relationship with regulatory practices was ambivalent. on the one hand, the practical
value of toxicological knowledge provided resources but, on the other, hand it could hinder the develop-
ment of more fundamental research lines. as in many fields of science there was a tension between
application oriented toxicologists and those with a basic research orientation (groenewegen 1988). This
tension can be illustrated by the two dominant ways to divide the field. The first distinguished medical,
environmental, industrial, pesticide and military toxicology, according to its application area. The second
distinguished biochemical, neurological and other approaches, according to the methods and theories
used (groenewegen 1988).
The tension also appeared in the struggle of toxicologists to establish a strong position in the biologi-
cal research council Bion. in contrast to most other fields there has never been a dedicated ‘Working
community’98 for toxicology. in 1981 a special committee (‘dwarsverbandcommissie’) was formed to
98 Working communities decided upon funding and research strategy/priorities within a specific field of biology, like ethology, and plant ecology and vegetation sciences.
128
chapter 5
stimulate toxicological research (groenewegen 1988). This was a bottom-up initiative from researchers
arguing that more research was needed into the mechanisms of toxicity and that the new field of ecotoxi-
cology deserved support. The name Dwarsverbandcommissie, which literally means connection-commit-
tee, indicates that toxicology was seen as a multidisciplinary field. it was also seen as an immature field,
not yet deserving its own Working group or community. The committee was not directly linked to
societal issues. indirectly, however, the practical value of the field was important to justify the initiative.99
The most important funder of academic toxicology has always been the government. in the mid
1980s, the government funded 90% of all toxicological research in the netherlands, and industry the
remaining 10% (W84). in 1987, the government introduced the program Stimulation of toxicological
research. it included a special budget for toxicology for four years. The goals of the program were to
coordinate research investments and to increase efficiency (t88). The government also reserved a bud-
get for organizing knowledge dissemination meetings (W90). The stimulation program focused on four
themes, all connected with practical applications: labor and health, development of new test methods,
eco-toxicology and toxicological compounds and chronic diseases in relation to nutritional habits
gradually a more basic research orientation developed within toxicology, for several reasons. The
shift was partly for practical reasons. already at the end of the 1970s, toxicologists at utrecht university
turned their focus to the molecular or cellular mechanisms of toxicity. ‘neurotoxicology, biochemical
toxicology and pathology were envisaged as providing fundamental support for environmental toxicol-
ogy and veterinary pharmacology’ (groenewegen 1988 p. 135). The design of the more fundamental
research was attuned to the problems in the practical studies, like the biochemical mechanisms of toxic
effects. This fundamental research, however, could in turn lead to new research lines, further away from
the original practical problems, developing into independent research projects with their own dynam-
ics, completely detached from the original problem definition. This shift in research content was also
visible in the research output. groenewegen’s analysis (1988) shows that between 1960 and 1980 dutch
reports and dutch journal articles became less frequent and gave way to international publications.
from 1983 onwards, simultaneous to the governmental program, research council Bion ran a priority
program with a more fundamental orientation (halffman 2003). Priorities of this program were funda-
mental issues such as combination toxicology and processes of poisoning and detoxification.100
Second, the expansion of more fundamental research lines in toxicology was stimulated by the grow-
ing pressure for scientific publications, catalyzed by the rise of performance assessments. it was encour-
aged to devote more effort to fundamental research lines that might result in publications in high-impact
scientific journals. asked about the importance of publishing a lot:
99 interview with former adjunct-director of Bion100 it was much smaller than the governmental program: only 200.000-500.000 fl. per year. The program was managed
cooperatively by the dwarsverbandcommissie for toxicology of Bion and the medical research council. There were several joint meetings with ministerial program and the two programs ended together in 1993 (halffman 2003).
129
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
Yes, that is what we are judged on! So ... that’s the visibility of the faculty, how productive one is, so to say.
So it is pretty important. (assistant professor)
illustrative is the fact that Phd students are nowadays required to publish about four papers during their
project. as in many other fields of natural science, Phd-theses are nowadays usually a collection of jour-
nal publications, rather than a monograph.101 Moreover, contemporary toxicologists strive for journals
with a high impact factor:
We actually have the demand that papers should be published in the top 30% of the field, but in fact, as an
institute, and we are succeeding reasonably well, the greater share of the paper in the top 10%. ... When I
am talking about toxicology, I almost always go for submission to the top 5 journals. ( full professor)
adding a medical context to one’s research can help to get into journals with higher impact factors:
So one has a nearly unlimited choice of journals. With different impact factors. Which all have a different
focus. The higher the impact factor, the better of course. And things with a disease simply sell better. But
then one has to position it in a different field. (assistant professor)
The third factor that changed the research agenda of toxicologists was the ‘problem’ that the most urgent
practical problems were solved, so there was less governmental funding available for applied research.
industrial toxicology, for example, hardly exists anymore as an academic field, because industry has
become cleaner, and there are now strict norms concerning exposure of industrial staff to toxic com-
pounds.102 as a result, toxicology remained linked with policy applications,103 but to be able to maintain
itself as a mature scientific field in a context of foresight and evaluation studies and to acquire sufficient
funding, toxicologists had to address fundamental questions as well.
in the mid 1990s, when both the ministerial research program and the Bion program ended, toxi-
cology was supported again by a Stimulation Program, focusing on the effects of chemical pollution on
ecosystems.104 This new program focused more on the mechanisms behind toxicity and was less policy-
oriented than the previous ones (halffman 2003).105 currently, there is no dedicated nWo program
for toxicology. in terms of funding, toxicology has never established itself as a mature field with its own
steady supply of money. The limited amount of funding for toxicological research that is available at the
research council nowadays solely comes from broader programs, requiring collaboration with other
101 interview with full professor in toxicology102 interview with nWo-alW representative103 in 1989 Setac europe is founded, a forum for science and policy making (halfmann, 2003)104 nWo-alW 1999105 confirmed in interview with nWo-alW representative
130
chapter 5
fields. The other organizations that may fund toxicologists are individual governmental ministries, the
eu, the medical research council ZonMw and medical charity foundations like the cancer foundation
and heart foundation. it is, however, difficult to acquire funding from any of these without collaborat-
ing with other fields. Both ZonMw and the medical charity foundations will only pay for projects that
are directed at a specific disease or patient group, and in most cases toxicologists cannot run such proj-
ects without the help of medical scientists. This serving role is illustrated with the following quote:
This Parkinson Foundation [project] is again a collaboration, in which we collaborate with a hospital in
Tilburg. There is a Parkinson department at that hospital. I happen to know the neurologist there. He had
some interest in pesticides and Parkinson. I had already written a couple of proposals for it. And we happened
to meet each other actually. And then the idea was to do it at the Parkinson Foundation. (assistant professor)
in conclusion, multidisciplinary research collaborations in toxicology have increased as a response to
external factors. They can both facilitate application oriented research and help to develop more funda-
mental research lines. in a sense, the need for multidisciplinarity has been present in toxicology since its
birth. The ‘dwarsverbandcommissie’ that was established in Bion to stimulate toxicological research
intended to form a bridgehead between different fields. notably, at utrecht university toxicological re-
search is still organized in an institute which is connected to the veterinary, the medical and the sciences
faculties. later on, the interaction with other fields served to give toxicological research more fundamen-
tal allure, as a response to societal pressure (in the form of performance assessments and funding pro-
grams) to develop a more academic research agenda. The tasks of exploring toxicity of compounds and
developing toxicity tests have been taken over by commercial laboratories and public research institutes.
to acquire funding for longer research projects, which yield Phd-theses and publications in high-impact
journals, questions on the mechanisms of toxicity on the level of organisms, cells or molecules need to
be addressed. These require collaborations with medical scientists or biochemists.
5.6 Conclusion
over the past few decades, the importance of multidisciplinary research collaborations in the life
sciences has grown. Multidisciplinarity is not a new phenomenon. in toxicology interactions with other
fields have been vital ever since its inception. Moreover, disciplinary structures are dynamic and cross-
disciplinary interactions can lead to the formation of new fields. Paleo-ecology itself can, for instance,
be seen as the product of collaborations between archaeology and ecology. however, the relative share
of multi- and interdisciplinary collaborations has grown significantly over the past 30 years (Porter &
rafols 2009).
131
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
The rise of multidisciplinary collaborations can be explained partly by internal factors, like the rise
of molecular techniques, improved measurement equipment and informatics (cassidy & radda 2005,
loging et al. 2007). While these developments are beyond the main focus of this paper, our interview
data indicate significant shifts in the dominant research systems over the past view decades. toxicolo-
gists increasingly work with cell-lines and computer models instead of test animals. Paleo-ecologists still
do a lot of individual microscope work, but they increasingly use molecular techniques as well. These
developments have probably increased the necessity of collaborating with other fields. Thus our case
studies confirm the significance of internal factors reported in earlier literature like personal contacts
(Melin 2000), mentoring relationships (Bozeman & corley 2004), specialization (laudel 2001) and
‘research ensembles’ (hackett et al. 2004) for the rise of multidisciplinary research collaborations.
Primarily, our analysis has identified three important changes in the ‘contract’ between biology and
society that have induced multidisciplinary research collaborations. The first development, which has
had a direct influence in both cases we have studied, is that multidisciplinarity has developed into an ex-
plicit policy goal, stimulated by various funding instruments. Second, society increasingly demands that
the life sciences deliver practical applications, which indirectly stimulates multidisciplinarity in paleo-
ecology (and to a lesser extent in toxicology, too). Third, the rise of quantitative performance assess-
ments has increased the pressure for scientific publications and academic excellence. This has indirectly
stimulated multidisciplinary research collaborations in toxicology.
academic toxicology was born out of a societal need and has used input from a variety of disciplines
from the start. under the current social contract it also uses multidisciplinary collaborations to enhance
its fundamental research. in response to the pressure from evaluations and funding instruments, toxicol-
ogy needs to improve the external validity of its approaches. Because standard toxicity tests are offered
by private labs and applied research institutes, academic toxicology now needs to focus on questions
concerning the mechanisms of toxicity to legitimate its existence. This requires collaboration with other
fields such as physiology and biochemistry.
Multidisciplinary research collaborations offer quite different benefits to paleo-ecology. in this field
the collaborations with other fields mainly serve to facilitate strategic research. for paleo-ecology the
changing social contract implied the need to develop a more application-oriented research line, con-
nected to climate change. Public policy has strongly supported collaborations with earth scientists that
are believed to yield new insights to inform environmental policy.
Before we close, let us discuss the generalisability of our findings. The findings reported in this chap-
ter are restricted to the netherlands. While specific conclusions about toxicology and paleo-ecology
should not be generalized to other countries, there are indications that the general changes observed in
the contract between biology and society have also taken place in other countries. Systematic perfor-
mance assessments are evolving all across western science systems (hicks 2009, Whitley & gläser
2007), strengthening the ‘publish or perish’ norm (Weingart 2005). Shifts in available funding sources
132
chapter 5
and selection criteria, and increased calls for practical relevance have also been reported elsewhere
(Slaughter & leslie 1997, Benner & Sandstrom 2000, geiger & Sa 2008). for this reason, our observa-
tion that the changing societal context of academic research can give rise to a variety of incentives for
multidisciplinary research collaborations can be generalized beyond the netherlands.
to conclude, our analysis of these two cases shows that multidisciplinary collaborations can serve
as means to different ends, and that the pressure exerted by the changing social contract with science
can result in different outcomes. This implies that generalized claims about the changing relationship
between science and society and the rise of ‘Mode 2 knowledge production’ (gibbons et al. 1994) are
severely vulnerable. The contract between biology and society has changed in various ways. it currently
exerts several forces on academic scientists that seem to be (partly) in contradiction, reminiscent of the
paradox around relevance observed in dutch chemistry: scientists are rewarded for making promises
about the (possible) relevance of their research, but not for realising these promises (see chapters 3 and
4). The funding sources available for the life sciences increasingly demand research proposals to show
the ‘relevance’ of the planned activities and to promise practical applications in the form of patents,
products or policy advice. at the same time, the pressure for scientific publications has grown, due to
the rise of systematic performance evaluations of university research. however, the relative influence of
these two trends varied across the two cases studied here. Multidisciplinary research collaborations ap-
pear a strategy to cope with the changing societal demands. toxicology, which started as an application
oriented field uses multidisciplinarity to become more fundamental, in order to cope with the pressure
for publications in prestigious scientific journals. Paleo-ecology, which started as an exotic, fundamental
niche in biology moves, through collaborations, to a more practical orientation that fits well with the
demands of current funding sources. in terms of Stokes’s quadrant model (figure 5.2) it seems that the
changing societal contract pulls different fields in the same direction: both fields use multidisciplinary
collaborations to move towards Pasteur’s Quadrant. Possibly the current contract between society and
academic biology is so demanding in terms of both practical applications and scientific output that
extreme positions are not sustainable. research fields in biology now need to combine basic and ap-
plied activities, in order to acquire sufficient funding and to score well in performance evaluations. for
researchers this is a delicate balancing act, and multidisciplinary collaborations appear as versatile means
to these ends.
133
Multidisciplinary collaborations in toxicology and paleo-ecology: equal means to different ends
Appendix B (chapter 5). Documents studied
Abbrevi-ation
Publisher / Author Year Title City
W74 Ministry of Science and education (oc&W)
1974 nota Wetenschapsbeleid The hague
B75-B93 research council Bion 1975-1993
Jaarverslag (annual report) The hague
W76-W97
Ministry of Science and education (oc&W)
1976-1997
Wetenschapsbudget The hague
ar79 academische raad 1979 Beleidsnota universitair onderzoek The hague
r80 Biologische raad ca. 1980
Biologisch onderzoek in de sub-fakulteiten biologie en instituten voor fundamenteel biologisch on-derzoek: rapport van de werkgroep ‘open Beraad’.
amsterdam
r83 Biologische raad 1983 Biologie van levensbelang amsterdam
r85 Biologische raad 1985 Bologisch onderzoek voor Mens en Maatschappij
amsterdam
K86 dutch royal academy of the Sciences (KnaW)
1986 Sub-disciplineplan Biologie amsterdam
t88-t90 dutch toxicological orga-nization (nvt)
1988-1990
toxpost
n92 research council nWo 1992 een organisatie op maat: Meerjaren-plan 1993-1997
The hague
S94-S98 research council SlW 1994-1998
Jaarverslag (annual report) The hague
v94 association of dutch uni-versities (vSnu)
1994 Quality assessment of research: netherlands Biology in the nineties
utrecht
n95 nWo 1995 trends in Wetenschap: achter-grondnota bij beleidsnota 1996-2001
The hague
M95 Ministry of agriculture, na-ture conservation and fis-heries; Ministry of Public housing, Spatial Planning and the environment; and Ministry of foreign affairs
1995 Strategisch plan van aanpak bio-logische diversiteit – nederlandse uitwerking van het verdrag inzake biologische diversiteit
The hague
134
chapter 5
vB97 verkenningscommissie Biologie
1997 Biologie: het leven centraal amsterdam
n97 nWo 1997 inspiratie en sturing van weten-schap: nWo en maatschappelijk georiënteerd en multidisciplinair onderzoek
The hague
a98-a02 nWo-alW 1998-2002
Jaarverslag (annual report) The hague
nJ99-nJ02
nWo 1999-2002
Jaarverslag (annual report) The hague
v99 vSnu 1999 Biology: assessment of research Quality
utrecht
W00 oc&W 2000 Wie oogsten wil, moet zaaien: Wetenschapsbudget 2000
The hague
c01 commissie disciplineplan Biologie
2001 Biologie: een vitaal Belang amsterdam
n01 nWo 2001 Thema’s met talent: Strategienota 2002-2005
The hague
g03 verkenningscommissie Bio-geologie
2003 tussen aarde en leven: een strategi-sche verkenning van de biogeologie in nederland
amsterdam
v03 vSnu, nWo & KnaW 2003 Standard evaluation Protocol 2003-2009 for Public research organi-zations
W04 oc&W 2004 focus op excellentie en meer waarde: Wetenschapsbudget 2004
The hague
v04 vno-ncW, vSnu & nfu
2004 Beschermde kennis is bruikbare kennis: innovation charter bedrijfs-leven en kennisinstellingen
e04 Ministry of economic affairs
2004 actieplan life Sciences The hague
v05 vSnu 2005 onderzoek van Waarde: activiteiten van universiteiten gericht op Ken-nisvalorisatie,
The hague
n06 nWo 2006 Wetenschap gewaardeerd! Strate-gienota 2007-2010
W07 oc&W 2007 voortgangsrapportage Weten-schapsbeleid
The hague
135
chapter 6.
137
Stakeholder interactions in dutch animal sciences106
AbstractThis paper investigates the effects of the changing institutional environment on academic research
practices in the case of dutch animal sciences. The two most important changes in the dutch agricul-
tural research system in the past few decades have been shifts in the available funding and the rise of
performance evaluations. our analysis shows that these have only stimulated interactions with societal
stakeholders in fields where this helped to sustain a basic research agenda. in other fields there turns out
to be a tension between satisfying the needs of application-oriented funding sources and reaching high
scores on evaluations that are dominated by bibliometric indicators.
6.1 Introduction
two common trends can be identified in the institutional environment of academic research in deve-
loped countries over the past 30-40 years: a shift from block-grant support to earmarked funding for
specific projects and programs (lepori et al. 2007), and the rise of performance evaluations which –
directly or indirectly – influence the availability of funding (hicks 2009, Whitley & gläser 2007). The
question to what extent these structural changes have transformed actual research practices remains
unsettled (hicks & Katz 1996). it has been claimed that academic researchers have become more ‘re-
flexive’, collaborate more intensively with users, and are subject to novel forms of quality control, which
reward societal contributions rather than scholarly products (gibbons et al. 1994, hemlin & rasmussen
2006). however, the available empirical data suggest that we should take care not to overestimate the
effects of institutional changes on research practices. dutch research departments often succeed to con-
tinue their existing research activities, even under profound institutional changes in their environment
106 This chapter was co-authored with John grin and ruud Smits and has been submitted for publication.
138
chapter 6
(leišytė et al. 2008). a study of american biologists suggests that the increase of industrial funding has
only subtle effects on scientific problem choice (cooper 2009). also in a case study of a finnish biotech-
nology department, research practices turned out to be remarkably stable, as it was difficult for academic
researchers to combine their university activities with commercial development (tuunainen 2005). in
some fields even a movement away from practical concerns has been observed, due to the increasing
pressure to publish in scientific literature (albert 2003).
against this background, the aim of our paper is to increase the understanding of the relationship be-
tween macro-level developments and micro-level research practices. By an analysis of the case of dutch
agricultural sciences we explore the ways in which institutional changes in a research system can lead to
changes in the way science is conducted and organized on the lab-floor. our analysis shows that institu-
tional structures do not determine academic research practices. The effects of institutional changes are
modulated by characteristics of scientific fields, such as existing research traditions and relationships
with societal stakeholders.
in this paper we focus on three fields of agricultural (animal) science as performed at dutch univer-
sities. as we explain below (section 2), we consider it important to differentiate between scientific fields
and between national contexts. dutch agricultural sciences have experienced a highly dynamic societal
context. over the past 50 years views on what agricultural research should be conducted, why it deserves
support, and how it should be organized have changed dramatically. contrary to popular notions such as
Mode 2 knowledge production, the main shift in the views on fields of animal science does not concern
an increasing pressure to contribute to practical applications. rather, these fields are under pressure to
contribute to other practical applications and to address different types of questions than before (grin
2010, grin et al. 2004).
The research questions guiding our analysis are the following:
1. What major changes have taken place in the institutional environment of dutch academic agricultu-
ral research since 1975?
2. What is the impact of these institutional changes on the daily work of dutch academic animal
scientists?
do these institutional changes stimulate academic researchers to interact more intensively with
societal stakeholders?
139
Stakeholder interactions in dutch animal sciences
6.2 Theoretical framework
to study the daily work of agricultural (animal) scientists, we use the ‘credibility cycle’ (latour & Wool-
gar 1986). This model explains how struggles for reputation influence the behavior of individual scien-
tists. its starting assumption is that a major motivation for a scientist’s actions is the quest for credibility.
conceived in this way, the research process can be depicted as a repetitive cycle in which conversions
take place between money, staff, data, arguments, articles, recognition, and so on (see also chapter 3).
But scientists do not work independently; their activities take place in the context of a ‘research
system’107. following rip and van der Meulen (1996), we regard a research system as consisting of
‘research performers (individuals, groups, institutions), other organizations and institutions, interac-
tions, processes and procedures’ (rip & van der Meulen 1996). This system contains universities, related
research institutes and funding agencies, but also governmental organizations, firms and intermediary
organizations to the extent that they are part of the institutional environment. This institutional environ-
ment provides research organizations with incentives and constraints to conduct (particular kinds of)
research. our notion of a research system is discipline-specific and it is delimited by national boundaries.
it is important to differentiate between scientific disciplines, because of their social and cognitive differ-
ences (Bonaccorsi 2008, Whitley 2000, Knorr-cetina 1999). as each discipline has its own (potential)
stakeholders and funding sources, the effects of institutional changes can be expected to vary across
disciplines (gläser et al. forthcoming). Moreover, literature about national innovation systems has
shown the importance of national science and innovation policies, national culture, and the presence of
particular economic sectors (lundvall 1992, nelson 1993).
in line with the structuration perspective (giddens 1984), the research system can be seen as the
structure influencing the agency of individual researchers (grin 2010). existing structures are the
product of practices and of dominant visions, such as the need to enhance agricultural productivity.
The institutions of this system give shape to certain conversions of credibility, e.g. the possibilities to
turn recognition into money (Packer & Webster 1996 , see also chapter 3). Simultaneously, funding
bodies presumably take into account the outcomes of research practices when formulating their future
priorities. in this way, research practices can strengthen these institutions, but they can also neglect them
and put them under pressure (Bos & grin 2008). So the research system can be seen as a structure that
shapes research practices, but that is at the same time (re)produced by these practices.
107 a research system can be seen as a subsystem of the national or sectoral innovation system. an innovation system is not only concerned with the development of new knowledge and technologies, but also with their adoption and diffusion (lundvall 1992, nelson 1993). So while (public) research organizations are central in a research system, in an innova-tion system they have to share this position with firms and other organizations that turn new knowledge into successful products, services or policy. furthermore an innovation system also contains boundary conditions for innovative entrepreneurship such as infrastructure, venture capital, intellectual property rights etc.
140
chapter 6
6.3 Methods
The research reported here is a set of case studies of three fields of agricultural science in The nether-
lands, all within the subset of (agricultural) animal sciences: animal breeding & genetics (aBg), animal
production systems (aPS), and cell biology (cB). We chose to study animal sciences because of their
relatively applied nature, which makes them an interesting set of fields to study the effects of the increas-
ing pressure for academically excellent performance (leeuwis 2000). Moreover, dutch animal sciences
have experienced a highly dynamic societal context, with various successive conflicts about issues such
as pollution, climate change and animal welfare, which facilitates the study of changing societal knowl-
edge demands (grin 2010). focusing on three fields of animal science enhances the possibilities for mu-
tual comparison, as they are all subject to the same contextual developments, i.e. the changing societal
views on and practices of animal production. Within this subset, however, these three fields represent
different ‘search regimes’ (Bonaccorsi 2008), characterized by convergence in aBg, and divergence in
aPS and cB. related, aBg has a steady relationship with a homogeneous set of knowledge users (ani-
mal breeding firms), while the other two fields have diverse stakeholders.
research question 1 was addressed based on document analysis, combined with expert interviews.
documents studied are listed in appendix c. The documents were collected based on prior knowledge
of the authors, tips from interviewees, and the ‘snowball effect’. The selection includes governmental
policy documents, reports and strategic plans of research councils, foresight studies, evaluations and
other important publications about Wageningen university. The findings from these documents were
triangulated in interviews with various experts directly or indirectly involved in agricultural science (see
table 6.1). our analysis of the changing agricultural research system is delimited to the period of 1975
until about 2005. The starting year of 1975 roughly marks the beginning of governmental science policy
in the netherlands (M74). in order to properly situate the history of agricultural science, however, we
provide an introduction to the agricultural research system that starts just after World War ii (section 4).
141
Stakeholder interactions in dutch animal sciences
Table 6.1 expert interviews108
name affiliation Position / expertise
Bram Bos Wur, animal Science group
expert system innovations
Johan Bouma Wageningen university Professor emeritus soil science, transforum advisory board
annemieke van der Kooij
nWo Senior program officer earth and life Sciences (alW), and Social Sciences and humanities (MagW)
henk van latesteijn transforum director, formerly deputy director of directorate knowledge of ministry of agriculture
niels röling Wageningen university Professor emeritus agricultural extension and innovation
gab van Winkel Wageningen university former secretary of research school WiaS
to answer research questions 2 and 3 we carried out semi-structured in-depth interviews with 12
academic researchers in the selected fields. The respondents’ ranks range from post-doc researcher to full
professor. Most of them are employed by Wageningen university; two work at utrecht university (see
table 6.2). We have asked them questions about their current and past research activities, their personal
motivation, and their experiences and strategies concerning funding acquisition, publishing, scientific
reputation, and performance evaluations. The questions had an open character, in order to minimize
our influence on the answers and to benefit optimally from the respondents’ own experiences. Because
of the sensitivity of some of the issues addressed, quotes from the interviews will be presented anony-
mously, and all interviewees will be referred to as ‘he’, regardless of their sex.
Table 6.2 interviews with academic researchers: distribution of respondents over sub-disciplines, universities and academic ranks
animal production systems (4)animal breeding and genetics (4)cell biology (4)
Wageningen university (10)utrecht university (2)
full professor (3)associate professor (2)assistant professor (4)Post-doc researcher (3)
108 Wur: Wageningen university and research centre (further in this paper denoted as au), nWo: dutch organization for Scientific research
142
chapter 6
6.4 Introduction to the Dutch agricultural research system
dutch agricultural research got a boost in the period after WWii, when the dutch government invested
a lot of money in it with the aim to ‘modernize’ the agricultural sector in order to increase its productiv-
ity (grin 2010). it is generally assumed that a key factor in the impressive productivity growth of the
dutch agricultural sector in the 1950s, 1960s and 1970s was the exceptional degree of consensus about
the direction of change. Politicians, farmers and researchers all shared the aim of increasing agricultural
productivity and agreed that mechanization was the key to reach this goal (Bieleman 2000, roseboom
& rutten 1998). Thanks to the strong alignment of the different participating actors, the research system
of this era has become known as the ‘ovo-triptych’, the strongly interlinked combination of research,
education, and extension activities (Maat 2003). Within this configuration (see table 6.3), a classical ex-
ample of the linear innovation model, the agricultural university (au)109 was concerned with more fun-
damental research and the applied research was located in separate institutes and in a number of regional
‘experimental stations’ (Wr91)110 (Schot & van lente 2003). The national council for agricultural
research (nrlo) functioned as an intermediary to coordinate the activities of the various research
organizations (dijksterhuis & van der Meulen 2007). The alignment of research agendas with practical
knowledge needs was further strengthened by the fact that most agricultural researchers originated from
farming families.
academic agricultural research was financed mainly by lump-sum funding from the Ministry of ag-
riculture and fisheries (Maat 2001)(Wr91). researchers at au received unconditional support, as the
usefulness of their work was not disputed. Because of the strong bonds between their (basic) research
activities and agricultural extension, it was taken for granted that their research was attuned to the needs
of farmers (and related industries) and that the results of their work would diffuse to the individuals or
organizations that could apply them.
109 until 1986 the agricultural university was known as landbouwhogeschool Wageningen. it was founded in 1876 as the ‘rijkslandbouwschool’, and became a university for professional education in 1918. Since 1986 it has the status of a research university. for the sake of clarity, we will use au throughout this paper.
110 codes in brackets like this refer to the documents listed in appendix 1.
143
Stakeholder interactions in dutch animal sciences
Table 6.3 overview of the organizations of the agricultural research system and their relationships with the agricultural university (au), around 1975
organization relationship with au
Ministry of agriculture and fisheries (Mlv)
Main funding source, in the form of block grant support
research council for Pure Scientific research (ZWo)
additional funding for fundamental research on competitive basis
extension Service Knowledge transfer to farmers
institutes for applied research attunement of research agenda
experimental Stations attunement of research agenda
national council for agricul-tural research (nrlo)
coordination of research activities of different organizations by keeping a project database and stimulating contact with applied research institutes and experimental Stations
after a period of impressive productivity growth, this research system gradually came under pressure
since the 1970s. concerns about the side effects of agricultural modernization led the mainstream con-
viction that the impact of the high-intensive agriculture on nature, on the environment and on the dutch
scenery was unacceptably high (Bieleman 2000). in addition, there were concerns about animal welfare
and overproduction and it became increasingly clear that on the long term, dutch agriculture would
not be able to compete internationally on price, because other regions possess comparative advantages
in terms of the potential for scale enlargement and the costs of land, labor and energy (van der Ploeg &
ettema 1990). to overcome these major problems, a radical change seemed necessary from mass pro-
duction of agricultural goods to sustainable production of knowledge-intensive specialties (Smits 2002).
The crisis in the agricultural sector implied that other kinds of knowledge were needed from the
agricultural research system. in 1982, when the responsibilities of Mlv were expanded in the areas of
recreation, nature conservation and environmental issues, nrlo also broadened its scope (dijksterhuis
& van der Meulen 2007). in addition to knowledge about how to enhance the efficiency of existing
systems of mass production, a need was developed for knowledge about animal welfare, about the envi-
ronmental impact of agricultural activities and about alternative production systems.
6.5 Institutional changes in the research system
The agricultural research system has undergone major changes over the past few decades. in short, one
can observe a shift from the consensus-driven ‘ovo-tryptich’ to a heterogeneous agricultural research
system. This shift consists of two major changes in the institutional environment of academic agricul-
144
chapter 6
tural research: shifts in the available funding, and the rise of systematic performance evaluations. These
will be subsequently addressed in this section.
Shifts in available fundingover time, the relative share of governmental block-grant support has decreased. an increasing share
of all funding available for agricultural researchers comes from projects and programs, paid by a variety
of organizations. While the former left considerable autonomy to the researchers regarding the content
of their work, the latter are often dedicated to making specific contributions to a more sustainable and
competitive agricultural sector.
The 1980s brought increasing scarcity of funding and increasing programming of research activities.
for the agricultural university (au), the governmental operations called ‘task division and concen-
tration’, and ‘Selective Shrinkage and growth’ implied budget cuts of 6.7 Million and 4.5 Million dfl,
respectively (faber 1993). in relation to the number of students, the governmental block-grant support
decreased with about 56% in real terms (faber 1993). Moreover, with the introduction of conditional
financing in 1983 (Blume & Spaapen 1988), a substantial share of the remaining lump-sum funding
became subject to programming and evaluation.
at the same time, the emphasis in the available funding shifted towards application oriented
research. in the 1980s the national research council was reorganized and started to fund application
oriented research as well111 (Kersten 1996). Meanwhile contract research became increasingly signifi-
cant for au112. in 1981 a ‘transfer point’ was established, to stimulate and facilitate project acquisition
and knowledge transfer (faber 1993). The volume of contract research grew from 12 Mdfl in 1982 to
74 Mdfl in 1991. around 1990 about 20% of all income of au stemmed from contract research (faber
1993).
dutch agricultural sciences faced a further rise of performance-based funding and earmarked pro-
grams in the 1990s. Block-grant funding for research gradually gave way to more competitive funding
arrangements (r96c). This general move from lump-sum funding to market-oriented funding took place
in all western science systems (levidow et al. 2002, huffman & Just 1999)(i03) and fitted in neo-liber-
alism, or the policy of ‘steering at a distance’ (de Boer et al. 2007). in the period between 1978 and 1995
the number of full-time equivalent research positions paid from block-grant support increased from 223
to 361, while those paid from other sources rose from 77 to 472. in 1995 the Ministry of agriculture
was responsible for less than half of all research income at Wageningen university (roseboom & rutten
1998). at the same time, the Ministry also privatized the agricultural extension service and its applied
111 Moreover Stichting technologische Wetenschappen (research council for technical sciences) was founded in 1980.112 The perception of contract research had radically changed since the 1970s, when contract research was considered
‘not done’. in the 1970s influential left-wing student groups fiercely protested against collaborating with industry. in 1972 one of the parties stated in its election programme for the faculty council that it aimed to ‘eliminate the contract research for the military-industrial complex’ (faber 1993, p233).
145
Stakeholder interactions in dutch animal sciences
research institutes (Maat 2001), which implied a breakdown of their bridging function between farmers
and au.
to a smaller or lesser extent, many of the agricultural research programs that have substituted the
lump sum funding express the need for sustainable development. after a turbulent phase, in which
there was little consensus about the desired direction of dutch agriculture and about the contribution
of agricultural research, views began to converge in the 1990s around the concept of sustainability113.
although it has not completely replaced productivity as the central goal of the agricultural research sys-
tem, sustainability has acquired a prominent position in arguments to defend investments in academic
agricultural research114.
in the course of years, an increasing share of research funding is governed by network structures
rather than top-down steering (Klerkx & leeuwis 2009, Klerkx & leeuwis 2008b). The concept of the
‘ovo triptych’ has been gradually replaced by ideas like ‘knowledge network’ (M93), ‘knowledge sys-
tem’ (r98a), ‘innovation system’ (Wr91) and ‘knowledge infrastructure’ (P96), which emphasize the
importance of feedback mechanisms and interaction between users and producers of knowledge, and to
include a wider variety of users that goes beyond the traditional agricultural stakeholders.
Since the mid 1990s a new feature has entered the knowledge infrastructure: consortia in which
public and private parties together finance a research program, partly conducted at Wageningen uni-
versity. in 1997, for instance, the Wageningen centre for food Sciences was founded, a collaboration
of several knowledge institutes (both within and outside Wageningen university) and food firms to
carry out basic research and to generate knowledge for the dutch food industry. another example is
the research and innovation program transforum, which started in 2004 and which qualifies itself as a
‘new knowledge arrangement’, with a mandate to contribute to innovation and transition, and with the
explicit intention to realize a transformation in the knowledge infrastructure, to make it more demand-
driven and interdisciplinary (t07)(hoes et al. 2008).
another significant development regarding the available funding for academic research concerns the
increasing emphasis within the national research council on stimulating individual talent, embodied by
the ‘vernieuwingsimpuls’ grants. Since its introduction in 2000 its relative share in nWo’s total budget
has increased to about 20% (n10). in the selection of proposals for these grants, the most determining
factor is the individual quality of the requesting scientist, assessed mainly using bibliometric criteria.
113 ‘Sustainable development, “sustainable” in ecological, economical, social and spatial sense, constitutes a central point of attention in the lnv-knowledge policy’ (M93). ‘The agricultural university wants to develop and disseminate scien-tific knowledge which society needs to meet its demands for sufficient and healthy food and a good living environment for human, plants and animals in a sustainable way’ (W92). ‘in all sectors sustainability is becoming a self-evident and at the same time challenging part of (scientific) practice’ (r96c).
114 nowadays the notion of sustainability is mentioned in many documents dealing with agricultural research, but not often as its central goal. according to B06 ‘sustainable entrepreneurship’ is one of the goals of agricultural research and ‘Sustainable agriculture and fisheries’ is an aspect of one the three core areas making up au’s domain (W08).
146
chapter 6
Moreover, it is important to note that the money that was left to au increasingly became subject to
university research policies. Since 1992 au has allocated funding for Phd students to research pro-
grams, rather than to research departments (W92), which implied a loss of autonomy for departments
to choose their own research agendas. recently, au has undergone a substantial reorganization in order
to facilitate a stronger market orientation. in reaction to several critical reports (P96, r96c), the Minister
of agriculture decided in 1997 to merge its strategic research institutes115 with the agricultural university
into the Wageningen university and research centre (Wur). Since 2001, also the ‘practice-oriented’
agricultural research was integrated into Wur (B06). The result is a large matrix organization consisting
of five ‘Science groups’ (Plant, animal, agrotechnology & food, environmental, and Social), which
each comprise both (fundamentally oriented) university departments and applied research departments.
The aim of this new set up is to facilitate knowledge exchange between the fundamental and applied
departments within each Science group, but also to create synergy between their funding acquisition
activities.
to conclude, in the complex dynamics of the funding sources of agricultural research, two trends
are visible. first – and most significant - an increasing share of the funding is dedicated to projects that
aim to make specific contributions to a more sustainable (or competitive) agricultural sector. applica-
tion oriented contract research, earmarked research programs and consortia, often including stakeholder
interactions, now constitute a substantial part of all available funding. Second, the personal grants of re-
search councils nWo and StW, which reward individual researchers with excellent publication records,
have become a significant source of funding.
Rise of performance evaluationsas other fields of science, the agricultural sciences became subject to performance evaluations during
the 1990s. in the 1970s research outcomes were mainly evaluated on the level of chair groups. only very
costly or long-lasting projects needed to be accounted at the faculty level (faber 1993). in addition, the
national council for agricultural research (nrlo) monitored the research activities at au, but this
was oriented at mutual attunement of these activities with the components of the ovo-triptych, rather
than assessing their performance (dijksterhuis & van der Meulen 2007). The increasing pressure on
available budgets and the growing need for accountability, however, have put ‘quality’ and ‘effectiveness’
on the agenda in the 1980s. in 1986 au started to experiment with systematic evaluations of research
quality and productivity (faber 1993). The need for systematic quality evaluations was shared by nrlo
(r86). in the early 1990s the association of dutch universities (vSnu) constructed a national system
for research evaluations of all scientific disciplines, which would also be used by au (W92). anticipating
this development, au also set up an internal quality control system, consisting of productivity analysis,
impact analysis and an assessment of the content by peer review (W92).
115 Known under the common heading ‘dlo’ (directorate for agricultural research).
147
Stakeholder interactions in dutch animal sciences
The animal science groups were assessed for the first time in 1999 as part of a national evaluation of
the ‘veterinary and animal Sciences’ (v99b), and later in the context of the evaluations of the graduate
school Wageningen institute of animal Sciences (WiaS) in 2004 and 2009. The first evaluation was
based on the second protocol of the association of dutch universities (v98)116, the latter two on the
new ‘Standard evaluation Protocol’ (v03). Both protocols prescribe a similar set of four main criteria,
namely quality, productivity, relevance, and vitality & feasibility. The most important indicators for the
first two criteria are citations and publication numbers. ‘relevance’ is a container concept referring to
both socio-economic and scientific impact. vitality and feasibility are measured in terms of manage-
ment and leadership of the program, funding acquisition, and ability to (re-)adjust the research program.
These evaluation reports give recommendations for improvement both on the level of the entire gradu-
ate school and on the level of individual research groups.
The research system around 2005as a result of the various changes just presented, today the agricultural research system is much more
complex than in 1975. Wageningen university now interacts with a broader set of organizations (see
table 6.4). The ministry of agriculture, nature conservation and fisheries is still the main supplier of
funds, but the relative contributions of other organizations have increased. other ministries (responsible
for economic affairs and environmental affairs), food and biotechnology companies, and non-govern-
mental organizations fund a significant share of the research activities. The research council ‘ZWo’ has
been replaced by the nWo, which – together with technology foundation StW – is less exclusively
oriented at basic research, and which supplies a large share of its money in the form of thematic pro-
grams. Moreover, new sources of funding are european framework Programmes (eu fPs) and public-
private consortia, such as transforum and the Wageningen centre for food Sciences. nrlo has been
reorganized into the innovatienetwerk (innovation network green Space and agrocluster), responsible
for conducting strategic foresight studies and for providing a breeding ground to ‘system innovations’.
116 interestingly, a sub-set of all agricultural sciences was evaluated with a novel methodology in a pilot study focusing ex-plicitly on societal impact (r99b), but it remains to be seen whether indicators of societal impact will get a prominent position in systematic evaluations.
148
chapter 6
Table 6.4 overview of the of the agricultural research system and their relationships with the agricultural university (au), around 2005
organization relationship with au
Ministry of agriculture, nature
conservation and fisheries
Main funding source, in the form of block-grant support and
research programs
other ministries funding for application oriented research
dutch organization for Scientific
research (nWo)
funding for fundamental and application oriented research
technology foundation StW funding for application oriented research
european framework Programmes funding in various forms
extension Service Knowledge transfer to farmers
institutes for applied research attunement of research agenda (within Wur)
experimental Stations attunement of research agenda
innovatienetwerk Strategic foresight studies; breeding ground for ‘system innova-
tions’
food and biotech industry contract research and consortia
environmental ngos funding for applied research, user committees, consortia
6.6 Relevance in practice: a credibility cycle analysis
in this section we will analyze the work of individual researchers, using the perspective of the credibility
cycle.
6.6.1 Animal Breeding and Genetics (ABG)
The institutional changes do not create any complications for animal breeding research. The rise of
performance evaluations has been an encouraging experience for researchers in aBg. in all three reports
this group received the highest score of the WiaS graduate school (v99b, W04, W09b). interview data
indicates that aBg has managed to adapt to the new demands of funding sources without compro-
mising its fundamental research interest and its effective research strategies that helped it to reach high
academic productivity. Thanks to the strong connections with animal breeding firms in combination
with the strong publication record, the Wageningen aBg-group has managed to maintain the continuity
149
Stakeholder interactions in dutch animal sciences
of research funding. it currently has an impressive portfolio of projects, supported by a variety of funding
sources117. Because its research has always been connected to animal breeding practice, the shifts in
available funding sources have not significantly influenced the degree of application-orientation of the
research agenda118.
The need to connect specifically with the dominant sustainability agenda has not required radical
changes in the research activities. Most of the new research questions are being addressed building on
available molecular and quantitative genetic approaches. historically, animal breeding research provided
tools and knowledge that firms needed for breeding more productive cows, pigs and chickens. currently
many of their activities are framed not in narrow terms of agricultural productivity, but in terms of a
better (or more sustainable) animal production system. This also concerns healthier meat, increasing
animal welfare, lowering the environmental impact etc119. Some projects dedicated to these new goals
were specifically designed accordingly and involve radically new activities that were not present in the
old research system. for example, one post-doc researcher aims to develop breeding methods that select
for chickens that are less aggressive and more ‘sociable’. This requires ethological observation of animals,
which traditionally was not conducted within the aBg group. however, many other researchers contrib-
ute to the new goals by further developing rather than replacing existing research techniques120.
in spite of this continuity in the research activities, the shifts in available funding have stimulated
aBg researchers to intensify the existing interactions with breeding firms and with the more applied re-
searchers of the aBg division in lelystad. The aBg divisions of au and of the applied research institute
in lelystad have even issued a shared promotion flyer to advertise their services. Moreover, together they
organize a ‘round table meeting’ every two years for all dutch researchers and practitioners involved in
animal breeding and genetics for networking purposes. contacts with the firms help to attract research
funding, both from the firms themselves and from public sources such as StW. animal breeding firms
are relatively large knowledge-intensive companies employing a lot of aBg’s graduates. They engage
in various types of collaborations ranging from bilateral collaboration projects, funded entirely by a
company, to public-private consortia and StW-funding in which a company’s interest is a necessary
requirement, but need not be expressed in financial terms. in the latter case, firms participate in a ‘users
117 The group now consists of over 50 researchers, including Phd students (http://www.abg.wur.nl/uK/staff, accessed on february 18th, 2010)
118 one interviewee even argues that aBg’s research has become more fundamental due to the growing size of research programs and the need for further specialization: ‘first it was pretty much the application [which dominated] and also a bit of the foundation. and over the course of time the application has not moved ou of sight, but we are more intensively in the backgrounds.’ (researcher 7).
119 The brief program design of aBg’s quantitative research division states: ‘Balanced selection programmes must be developed and implemented that respect the genetic variation in adaptive capacity of animals, as expressed in their re-sistance to disease, their reproductive rates and welfare.’ (http://www.abg.wur.nl/uK/research/Quantitative, accessed on february 18th, 2010)
120 interviews 5 and 7.
150
chapter 6
committee’. They are updated regularly on the progress of the project and get the opportunity to provide
suggestions regarding the direction of the research.
The interactions with breeding firms appear to have some influence on the general direction of
aBg’s research activities, but they do not strongly restrict the researchers’ autonomy121. for this reason,
the increase of industrial funding does not imply more applied research. Moreover, interactions with
breeding firms offer two other benefits, beside access to funding. first, the interactions with firms
stimulate the researchers by providing challenging research questions and by turning fruitful research
outcomes directly into practical innovations. all aBg researchers experience the direct relevance of the
work for breeding companies as a motivating factor122. Second, firms provide access to valuable research
data. Their large populations of animals enable university researchers to work with much larger datasets
than they could do when they had to breed and keep all the animals themselves.
The high evaluation scores (v99b, W04, W09b) suggest that the increased publishing pressure stem-
ming from performance evaluations is not problematic for the aBg group. asked whether the pressure
to publish has increased over the past 30 years, a senior researcher replies:
‘That has always been the mainstay of everything. Has it increased? Well, there is certainly more attention
for it. And of course now it is all more... Thanks to things like Web of Science it is increasingly easy [to meas-
ure publications] and that is also important, because I don’t see it as a burden. Good work is wonderful,
but if it is not being written down, you might just as well not have done it. And here we do have a public
mission, we don’t have the attitude of: enjoy yourself. I would say: enjoying yourself is wonderful, but do
share it with others.’ (researcher 7)
The fact that the interviewee is not even sure whether there is currently a higher pressure to publish,
indicates that he does not experience this pressure as threatening. The group manages to publish papers
on all types of research projects, regardless of their funding source. interviews with other researchers
of this group123, however, clearly indicate that more junior researchers do experience a strong pressure
to publish. They regard producing many papers, preferably in high-impact journals, as crucial for their
future career, either within our outside the aBg group.
121 ‘Well, in general they fairly tend to give us a free hand in order to see where that leads to and what it yields.’ (researcher 6)
122 all four aBg researchers we have interviewed referred to the possible practical applications of their work by breeding firms when asked about their motivation.
123 interviews 5 and 6.
151
Stakeholder interactions in dutch animal sciences
6.6.2 Animal Production Systems (APS)
from the perspective of the changing societal knowledge demands, the aPS research seems very timely.
it addresses the need for research on sustainable agriculture and is relatively strongly oriented at practical
applications. however, in this field interactions with stakeholders are not always productive in terms of
scientific publications. There seems to be a contradiction between the increasing pressure for academic
publications and the need to address questions of societal stakeholders.
The scores of aPS on the three performance evaluations have been relatively low (v99b, W04,
W09b). The most recent evaluation committee judged that ‘the societal and cultural impact of the work
was excellent but the citations and relative impact were relatively low’ (W09b, p39) and it also com-
mented that the volume of publications should increase. These evaluations are taken very seriously by
the management of the graduate school and the faculty, and the scores can influence their strategic deci-
sions124. Some aPS researchers feel torn apart by the difference between criteria used in performance
evaluations (and in local management decisions) and the demands of their funding sources:
‘If you work 4-5 years on something and afterwards the conclusion is: we have contributed little to the
livelihood of the local population, then that’s is not good either. That is always a sort of clash between
multiple interests. You will be judged on whether you will have delivered your work in four years like it has
been promised as well, isn’t it? We will also be judged on hard criteria just the same. But if you want to get
projects again, for example in [the] INREF [program] or at [the] WOTRO [department of NWO], then
you also have to come up with other criteria. So it always remains a kind of giving and taking between
formal criteria that are imposed on you by your boss, so to say, and the criteria that others impose on you.’
(researcher 1)
however, another interviewee (researcher 15) argues that the practical impact of the aPS activities
is also appreciated by the wider Wageningen animal science community. Possibly more than at other
universities, researchers at au also give recognition to their colleagues for practical successes, even if
they have not yielded impressive academic publications. This probably relates to the intrinsic motiva-
tion many of them feel to contribute to the development of agricultural production, in the netherlands
or abroad125. This suggests that the informal reputation of a researcher (or research group) can still be
enhanced by contributions to human welfare or sustainable agriculture, but the formal reputation that
determines funding decisions of university managers is dominated by bibliometric criteria.
124 ‘This year there is an assessment again, and its results will be pretty important this year, because it actually determines the opportunities you will get in the future. Your research is being managed through research schools. and if you get a low assessment, then that will have an impact on what you can do in the future.’ (researcher 1)
125 all researchers in aPS and aBg that we asked about their motivation referred to the possible applications of their work in agricultural practice. researchers in cB turned out more motivated by a quest for fundamental understanding.
152
chapter 6
With regard to money, aPS depends to a large extent on funding sources with a practical orientation.
The opportunities at nWo are limited, due to aPS’s interdisciplinary and application oriented research
portfolio.
‘The final assessment at NWO, I always feel, is still very much based on the old norms. Then they say: “a
great interdisciplinary project, but we don’t have a clue what you are going to do”. I also notice that I am
often evaluated by people lacking interdisciplinary or application oriented education.’ (researcher 12)
a further limitation to the access to nWo funding is the problem that few of the aPS researchers have
an exceptionally strong publication record.
The experiences with practically oriented projects vary, depending on their duration. one research-
er, who is involved in many short research projects, feels that he is forced to engage in consultancy work
rather than in-depth research that is suitable for academic publications. But another, who carries out
more substantial projects, argues that practical and scientific goals can be fruitfully combined:
‘Well these people are paying you, so they do have a certain interest. And so the challenge is to... then you
make a proposal and when you go there to present it.... the challenge is to position it in such a way that you
very clearly sketch their interest, but that you’re still able to do what’s important for you to do. That is the
challenge. And then you end up in a win-win situation.’ (researcher 3)
aPS intensively interacts with ‘agricultural practice’, in various forms. These interactions help in gaining
credibility, but not in such a fruitful way as in the aBg group. aPS researchers have always had contacts
with farmers as part of the research strategy. The aPS-approach involves data collection in the form of
monitoring or experiments at individual farms. other interactions serve as part of funding acquisition,
and they have become more important due to the decrease of block-grant support. animal feed industry
and applied research institutes such as gezondheidsdienst voor dieren and animal Science group le-
lystad are of increasing importance as sponsors of the research activities of aPS. however, the collabora-
tion with animal Science group lelystad is much less intensive than in the case of aBg.
Publishing sufficient numbers of articles in scientific journals seems more difficult in aPS than in
the other two fields, for several reasons. The first is the relatively practical orientation of many research
projects. They often take place in the global south and serve local development goals, the relevance of
which for scholarly debates is not self-evident. The second problem is that some of the funding comes in
very small portions (e.g. the projects for the Science Shop). combining the outcomes of several projects
153
Stakeholder interactions in dutch animal sciences
in order to produce a scientific paper is quite challenging and requires additional time which is difficult
to get funded126. The third factor complicating scientific publishing is the fact that aPS does not belong
to an established discipline with a strong research tradition127. The aPS approach combines elements
of various schools, such as systems research, environmental science and development studies. for this
reason there are few journals available as a stable platform for publishing results. aPS researchers have
numerous frustrating experiences with peer reviews of their manuscripts, in which they feel being evalu-
ated with inappropriate standards. The development of a visible research tradition is complicated by the
heterogeneity of aPS’s research activities. as the group combines research in tropical areas with research
on dutch agriculture, using a variety of methodologies, the group’s members each have their own
networks and audiences and they do not collectively contribute to the foundation of a new disciplinary
tradition.
6.6.3 Cell Biology
of the three fields under study, cell biology has the strongest tradition of fundamental research. When
compared to aBg and aPS, the research activities in agricultural cell biology are less strongly connected
to practical applications: ‘it is fundamental, but also with a wink to applicability. it’s not that we say: if we
are going to investigate this, we can apply it already within four years. That’s not how it works. it is more
like: we understand how it works, and we also understand of certain diseases what kind of molecular
foundation it has. and then, once we understand it, we can start looking for a solution. That is step two.’
(researcher 11) The primary interest of these activities is to enhance the fundamental understanding of
biological mechanisms on the cell level. on the longer run, this understanding can contribute to more
effective (or sustainable) fish production and to improved medical therapies to various diseases, either
for humans or for animals.
although the cB group in Wageningen scored high in the 2009 evaluation (W09b), it is generally
difficult for the groups in cell biology to reach impressive scores on performance evaluations, because
they have limited access to journals with a high impact factor. These journals are dominated by research
with a medical orientation and publish few studies on agricultural animals:
‘A clinical immunologist, or a medical immunologist who is working on mouse, often can reach much higher
journals and therefore reach higher citation indices than a veterinary immunologist who is part of a much
smaller scene. A veterinary immunologists is already happy with attaining an impact factor of 1.5. An
126 for this reason one researcher (with an impressive publication list) chooses not to engage in projects that are so short or applied that they do not provide good perspectives on scientific publications in the first place.
127 The aPS group also mentions the difficulty to publish interdisciplinary work in their SWot analysis for their peer review evaluation (W09b, p39).
154
chapter 6
immunologist at a medical faculty or a biomedical faculty is not going to do it for a paper in a journal [with
an impact factor] under 7.’ (researcher 9)
for similar reasons, the agricultural cell biologists have a hard time getting funding from the research
council nWo. at the life sciences department they have to compete for grants with researchers with a
more fundamental or medical orientation and a more impressive publication list. apart from the relative
disadvantage in terms of publications, researchers also complain that referees and program committees
of the research council do not consider research done on agricultural animals of much added value. out-
side nWo it is also increasingly difficult for cB to find funding. unlike aBg, cB does not have a steady
relationship with a particular industrial sector, as the possible applications of its research and the labor
market of its graduates are quite diverse128.
in order to overcome this situation, the two groups pursue different strategies. The cB group at
utrecht university is gradually shifting its activities to more basic research, in order to publish more in
high-impact journals which should help to get more nWo funding (or similar eu-grants) on the long
run. although embedded in a veterinary faculty, the research is increasingly concerned with human
medicine. The Wageningen cB group spreads its activities over various themes, combining fish, livestock
and human cell biology (and immunology), in accordance with the available funding. focusing too
much on human or medical research is not legitimate within WiaS, but it seems necessary to a certain
extent in order to reach sufficient academic productivity. The following quote from the most recent
evaluation report illustrates that the human cell biology research line is not considered immediately ap-
propriate, but does have strategic value:
‘The human research does not have a good ‘fit’ within WIAS. However, this is the main area of interest for
the Chair, the work is externally funded and so the costs to WIAS are relatively modest, and there are tangi-
ble benefits for the remaining programmes, helping these to stay at the forefront of the field.’ (W09b, p41).
The cB researchers do not interact intensively with societal stakeholders. The utrecht interviewees re-
port hardly any interaction outside the scientific community at all. The researchers in Wageningen have
regular contact with firms that (co-)fund some of their projects or that participate in users committees
or feedback groups of projects funded by public sources such as StW or ZonMW. These interactions
are limited, however, and the 2009 evaluation committee commented that the ‘downstream potential’ of
the cB research should be exploited better (W09b, p41).
128 The basic research tradition of the cB group in Wageningen can be explained by its origin. The group was founded in the early 1970s when a new education program in (general) biology was set up. as this education program did not have a specific agricultural orientation, scientists were hired with a fundamental biological background (interview researcher 9).
155
Stakeholder interactions in dutch animal sciences
although all cB researchers experience a strong pressure to publish sufficient papers, their reactions
to this pressure vary, depending on their personal ambitions. one postdoc researcher argues:
‘Personally, I don’t have a strategy to publish as much as possible. I simply do my job. I don’t want to write
as many papers as possible, I find it more important to deliver sound research.’ (researcher 8)
But he admits that this stoicism is probably not beneficial for his future career. another postdoc shows
opportunism in his reaction:
‘Lately I have focused pretty much on minimal publishable units. (...) I think you have to keep your eyes
open very well: is this already publishable or not? Once you know sufficient, what kind of journal would be
suitable, can it already go? That is sort of my latest strategy.’ (researcher 10)
he pursues this strategy in order to avoid creating ‘gaps’ in his publication list, which may inhibit his
chances on a permanent research position.
6.7 Discussion
The aim of this paper is to increase the understanding of the relationship between macro-level changes
in the governance of science and research practices in different scientific fields. We have observed two
profound changes in the institutional environment of academic agricultural research in the netherlands.
dutch agricultural science used to be part of a stable and smoothly running system, oriented at the en-
hancement of agricultural productivity through modernization. The relevance of the academic research
was self-evident, thanks to the strong connections between research, extension and education activities.
Since the 1970s, however, the available funding has shifted, and as a result a greater share of the available
money has become dedicated to research that promises to make specific contributions to a more sus-
tainable and competitive agriculture. at the same time, performance evaluations have developed into
a systematic element of the research system, and provide input to decision making both by university
managers and by external funding sources.
our credibility cycle analysis shows that these developments have two generic consequences for
daily research practices in all three fields of animal sciences we have studied (figure 6.1). first, perfor-
156
chapter 6
mance evaluations intervene in the conversion of articles into recognition. Their main consequence is
that recognition is formalized and expressed in numerical values. evaluations give scores to the work car-
ried out by research groups. although these are not the only indicators used, bibliometric indicators, like
the numbers of publications and citation scores are the most visible, tangible and objective indicators.
The development of indicators for systematically measuring social or economic impact is currently still
in the pilot-phase (Spaapen et al. 2007). Besides, such societal indicators, which are measured to some
extent under the criterion ‘relevance’, are usually not something to worry about in animal science. Most
of the research lines are so strongly connected with agricultural practice that they are almost relevant by
definition129. for these reasons, the bibliometric indicators tend to dominate in the interpretation of the
evaluation reports, and in the discussion of their implications. as a result, bibliometric indicators are
increasingly used to express a researcher’s quality, also on the individual level. in other words, research-
ers earn recognition increasingly based on a quantitative perspective of their output:
‘This pressure [to publish] is self-imposed. But if you do not do this, and you work for a longer period with-
out projects, and without additional support from PhD students, postdocs, then at a certain moment you
will contrast sharply with an other. Even if you are pretty good.’ (researcher 8)
as this quote illustrates, the content of a researcher’s papers matters much less than the number of
papers, number of citations, the ‘impact factor’ of journals one publishes in, or even the so-called hirsch-
index130.
The second generic effect concerns the way recognition can be turned into money is changed by
the shifts in available funding sources and their respective demands. in the 1970s research groups could
count on sustained financial support, as long as they performed more or less adequately. au supplied
a basic level of funding for permanent staff, but also for some junior researchers and the equipment
required to conduct a number of research projects. researchers could attract additional funding from
external sources, but this was not necessary for survival. nowadays the block-grant support is hardly
enough to finance the permanent staff of a research group. in order to secure their own position,
researchers need to continuously write proposals and negotiate with stakeholders that may be willing
to pay for contract research. This development implies that recognition from only a limited group of
scientific peers is not sufficient for the acquisition of funding. one needs to convince a broader set of
stakeholders, outside university, of the value and viability of one’s research agenda. it has become more
important to explicitly promise a practical impact. in particular, researchers need to align their plans
129 This is illustrated by the documentation WiaS has prepared for the 2009 evaluation (W09a), which is a 320-page report consisting only of scientific publication lists and academic cvs. it does not address the economic or societal impact of the WiaS research at all.
130 in Wageningen an ‘h-index dinner’ with the university president seems to have taken place in 2009, for which the researchers with the highest hirsch-index of each department were invited.
157
Stakeholder interactions in dutch animal sciences
with dominant visions on the future of dutch agriculture. There are much more funding opportunities
for research that promises to contribute to a more sustainable and / or competitive agricultural sector
than for research that only promises to enhance fundamental understanding (or that promises a contri-
bution to the outdated agenda of productivity enhancement in the narrow sense). for research lines that
can not be directly connected to stakeholder demands, the only escape route seems to be the ‘vernieu-
wingsimpuls’ funding of nWo. This type of funding does not require promises of practical applications,
as the selection is mainly based on personal ‘excellence’. however, due to the strong competition, one
needs an extremely strong publication record to be eligible for these grants.
Figure 6.1 today’s credibility cycle of dutch agricultural science. The institutional changes have two generic effects on the credibility cycle, which are indicated in bold. at two other positions in the cycle, effects can be observed that vary across the three fields studied (indicated in italic).
But has the daily work of university researchers changed? do they interact more with ‘users’ of their
knowledge, in order to contribute to the agenda of sustainable agriculture? our analysis shows that
this differs strongly across fields. Both the fields aBg and aPS have increased their interactions with
stakeholders outside university, but these interactions are of a different nature. in aBg researchers have
frequent contact with breeding firms in order to select challenging research issues, to collect data, and
to acquire research funding. in aPS, however, contacts with farmers only contribute to data collection,
while contacts with applied research institutes and with ngos help to get access to funding. in cell Bio-
Pressure to publish
Funding instruments
Arguments
Data
Money
Recognition Articles
Performance evaluations
Interactions with users
Staf f and equipment
158
chapter 6
logy, interactions with ‘users’ have decreased rather than increased, as there is a tendency to shift towards
more fundamental research issues that are not of immediate interest to the agricultural sector.
in order to explain the differences across the fields three issues deserve attention. The first relevant
factor is the agreement between leading research questions and the organization of the field. in all three
fields a certain change in the leading research questions can be observed in the period studied, but the
degree to which the new questions are compatible with disciplinary infrastructures varies. The question
that is leading aBg today, i.e. how to breed animals for more sustainable agriculture, can be addressed by
approaches similar to the ones addressing the former question oriented at more productive animal hus-
bandry. The current activities can build on existing disciplinary infrastructures, such as scholarly journals
and scientific networks. for this reason, aBg can accommodate the institutional changes without pro-
foundly changing or reorganizing its own activities. The situation is different in aPS. in this field there
appears to be a mismatch between the societal knowledge demands and the existing disciplinary institu-
tions to support the fulfilment of these demands. The new leading question of designing and monitor-
ing sustainable agricultural systems requires a combination of knowledge dealing with animal feeding,
animal housing, and animal behavior, which have traditionally been separate. Moreover, the cooperation
of non-scientific actors such as farmers is also indispensible in aPS’s systems approach, but these do not
always see the immediate benefits of this work. This transdisciplinary approach is currently developed in
the aPS group but it is not yet supported by a clearly visible international research community, scholarly
organizations and established scientific journals. Journals with a respectable tradition (and a high impact
factor) stem from the era of the old research system, which was oriented at productivity enhancement.
The second important issue is the readiness with which scholarly contributions can be measured
by bibliometric indicators, in particular the journal impact factor. apart from a strengthening of mutual
competition, the application of bibliometric evaluations does not appear to have strong implications for
aBg. researchers have become more conscious of their bibliometric performance, but this does not
seem to have a significant impact on their choice of research activities. in aPS and cB, however, which
have more diverse and less coherent research portfolios, bibliometric indicators give uneven rewards
across different sub-fields131. in aPS studies on western agro-systems are generally more suitable for
publication in high-impact journals than studies on tropical areas. Similarly, high-impact journals in cB
favour mouse or human tissue as a research model over agriculturally relevant animals like chicken. This
situation can create a tension between research directions that are fruitful in bibliometric terms, and
directions addressing societal demands.
The third explanatory variable concerns the characteristics of societal stakeholders. aBg has an
unequivocal relationship with a homogeneous set of wealthy stakeholders. The knowledge-intensive
breeding firms take such a serious interest in fundamental knowledge (and in new generations of highly
skilled people) that they sponsor substantial research projects granting considerable autonomy to the
131 The lack of cross-disciplinary comparability is a well-known limitation of bibliometric analysis (leydesdorff 2008).
159
Stakeholder interactions in dutch animal sciences
scientists. for this reason, the increased interaction of aBg with animal breeding firms and with aSg
lelystad turned out not in conflict with the pressure for (more) scientific publications. Such a steady
relationship with a homogeneous group of stakeholders is lacking in cB and aPS. cB does not interact
intensively with any stakeholder group. The aPS group has increased its interactions with societal
stakeholders, but, due to the diversity of their interests, this may threaten the viability of the group. Most
projects for dutch ngos and for international charity foundations are too small and too diverse to form
the basis of a convincing contribution to scientific literature. aPS’s stakeholders appear not wealthy
enough to be able to afford investing in fundamental research projects that may only give returns on the
longer term. Moreover, aPS’s sustainable systems approach cannot count on univocal support of farmers
and agricultural organizations as it often challenges their current practices.
our findings regarding the institutional changes in the academic research system suggest that
scientific criteria still dominate in academic quality control. This confirms leeuwis’ suggestion that
applicability and practice orientation are not important evaluation criteria in the current reward system
of dutch agricultural science (leeuwis 2000, Klerkx & leeuwis 2008a). These observations do not
corroborate the shift to ‘quality monitoring’ that has been suggested by hemlin and rasmussen (2006).
our analysis does show an increasing importance of quality assessments in general, and also an increas-
ing intensity of interactions between university researchers and stakeholders, but not a uniform shift to-
wards more societal criteria and the inclusion of users in evaluation processes. in our study the inclusion
of stakeholders has become only visible in the allocation of research funding. Performance evaluations
turn out to be dominated by bibliometric criteria that stimulate an inward looking perspective in which
the role of societal stakeholders is marginal. This can be explained by the limitations of the available
indicators to evaluate social or economic impact132.
With regard to science and innovation policy, this study shows that the governance of science is
most effective when it adopts a long-term perspective. in the end, the unique quality of university
research remains producing fundamental insights, which will help to solve practical problems or create
economic benefits on the long rather than the short term (Kronjee & nooteboom 2008). addressing
new knowledge demands requires the availability of sufficient supportive disciplinary infrastructure. if
knowledge demands change much faster than the supportive infrastructure is able to adapt to, research-
ers will not be able to adequately address these demands. for new fields it is difficult to establish a stable
position, especially due to the high and continuous pressure to deliver excellent academic performance.
The case of aBg has shown that the most fruitful collaborations between academic research and societal
stakeholders develop when the latter are ready to invest in knowledge that will only yield profits on the
longer run. in fields with less generous and powerful stakeholders, scientists struggle to satisfy the needs
of application-oriented funding sources and still reach high scores on bibliometric evaluations.
132 This may change in the near future, as there are currently attempts to develop indicators to systematically analyze the societal impact of academic research as well (Spaapen et al. 2007).
160
chapter 6
Appendix C (chapter 6). Documents studied133
Abbre-viation
Publisher / Author Year Title City
W68 landbouwhogeschool Wageningen
1968 de landbouwhogeschool op een Keerpunt: Jubileumboek ter gelegenheid van het 50-ja-rig bestaan van de landbouwhogeschool Wageningen
Wageningen
r71 nrlo 1971 Waarheen met de structuur van het land-bouwkundig onderzoek?
den haag
r72 nrlo 1972 organisatiestructuur landbouwkundig onderzoek en achtergronden van haar totstandkoming
den haag
M74 Ministerie van onder-wijs en Wetenschap
1974 nota Wetenschapsbeleid den haag
a79 academische raad 1979 Beleidsnota universitair onderzoek den haag
r86 nrlo 1986 Bevordering van de kwaliteit van het land-bouwkundig onderzoek
den haag
i89 iSnar 1989 The agricultural research-technology interface: a Knowledge Systems Perspective
den haag
r89 nrlo 1989 rapport 89/32: commercialisering van kennis en het functioneren van het land-bouwkennissyteem
den haag
Wr91 Wrr / nrlo 1991 technologie in de landbouw: effecten in het verleden en beleidsoverwegingen voor de toekomst
den haag
W92 landbouwuniversiteit Wageningen
1992 de strategie richting 2000: Strategisch Plan landbouwuniversiteit Wageningen
Wageningen
133 This table only lists primary sources, which are not listed in the bibliography of this paper. abbreviations:iSnar: international Service for national agricultural researchKnaW: Koninklijke nederlandse akademie voor de Wetenschappen (royal dutch academy)nrlo: nationale raad voor landbouwkundig onderzoek (national council for agricultural research)nfu: nederlandse federatie van universitair Medische centra (federation of dutch academic medical centres)nWo: nederlandse organisatie voor Wetenschappelijk onderzoek (dutch organization for scientific research)ocv: overlegcommissie verkenningen (committee for foresight studies)vSnu: vereniging voor Samenwerkende nederlandse universiteiten (association of dutch universities)Wrr: Wetenschappelijke raad voor regeringsbeleid (Scientific council for governmental Policy)WiaS: Wageningen institute of animal Sciences
161
Stakeholder interactions in dutch animal sciences
r92 nrlo 1992 rapport 92/16: implicaties van het begrip duurzame ontwikkeling voor de neder-landse landbouw
den haag
M93 Ministerie van land-bouw, natuurbeheer en visserij
1993 lnv-kennisbeleid: eenheid in verscheiden-heid
den haag
Wr94 Wrr 1994 duurzame risico’s: een blijvend gegeven den haag
M95 Ministerie van land-bouw, natuurbeheer en visserij
1995 lnv-Kennisbeleid tot 1999 den haag
n95 nWo 1995 trends in Wetenschap: achtergrondnota bij beleidsnota 1996-2001
den haag
P96 Peper, Bram 1996 duurzame Kennis, duurzame landbouw: een advies aan de Minister van landbouw, natuur en visserij over de Kennisinfrastruc-tuur van de landbouw in 2010
r96a nrlo 1996 rapport 96/18 Kennisproduktie als weten-schap en praktijk: aard en verandering van de landbouwwetenschappen
den haag
r96b nrlo 1996 rapport nr. 96/15: essays voor de verken-ning “landbouwwetenschappen in 2010: de positie van de luW”
den haag / amsterdam
r96c nrlo/ocv 1996 Wageningen in profiel. landbouwweten-schappen in 2010: de positie van de luW
amsterdam
r97 nrlo 1997 rapport 97/17: uitdagingen en concepten voor toekomstig landbouwkennisbeleid
den haag
n97 nWo 1997 inspiratie en sturing van wetenschap: nWo en maatschappelijk georiënteerd en multi-disciplinair onderzoek
den haag
r98a nrlo 1998 rapport 98/1: een maatschappelijk perspectief voor de landbouw: Kennis- en innovatieopgaven voor de toekomst
den haag
r98b nrlo 1998 rapport 98/20: Kennis- en innovatieagenda: ambities voor de 21e eeuw
den haag
v98 vSnu 1998 Protocol 1998 utrecht
M99 Ministerie van land-bouw, natuurbeheer en visserij
1999 Kracht en Kwaliteit: het lnv Beleidspro-gramma 1999-2002
den haag
162
chapter 6
r99a nrlo 1999 rapport 99/1: Wetenschap en technologie: Kansen voor de agrosector, groene ruimte en vissector
den haag
r99b nrlo 1999 rapport 99/12: de evaluatie van universi-tair onderzoek: methodiek voor het incor-poreren van de maatschappelijke waarde van onderzoek
den haag
r99c nrlo 1999 rapport 99/17: innoveren met ambitie: Kansen voor de agrosector, groene ruimte en vissector
den haag
v99a vSnu 1999 assessment of research Quality: agricultu-ral Sciences
utrecht
v99b vSnu 1999 assessment of research Quality: veterinary and animal Sciences
utrecht
v99c vSnu 1999 Biology: assessment of research quality utrecht
nW01 nWo 2001 Thema’s met talent: Strategienota 2002-2005
den haag
v02 vSnu 2002 assessment of research Quality: chemistry and chemical engeneering
utrecht
i03 iSnar 2003 trends in the organization and financing of agricultural research in developed coun-tries: implications for developing countries
den haag
v03 vSnu, nWo and KnaW
2003 Standard evaluation Protocol 2003-2009 for Public research organizations
l03 landbouw-econo-misch instituut
2003 van ovo naar vovi: nieuwe institutionele arrangementen voor kennisverwerving en -ontwikkeling van agrarisch ondernemers
den haag
e04 Ministerie van econo-mische Zaken
2004 actieplan life Sciences den haag
o04 Ministerie van on-derwijs, cultuur en Wetenschappen
2004 focus op excellentie en meer waarde: We-tenschapsbudget 2004
den haag
v04 vno-ncW, vSnu and nfu
2004 Beschermde kennis is bruikbare kennis: innovation charter bedrijfsleven en ken-nisinstellingen
W04 WiaS 2004 international Peer review of the WiaS graduate School
Wageningen
163
Stakeholder interactions in dutch animal sciences
B06 Berenschot 2006 groene Kennis (de)centraal? evaluatie van de Wijzigingen in het landbouwkundig onderzoek
utrecht
n06 nWo 2006 Wetenschap gewaardeerd! Strategienota 2007-2010
den haag
t07 transforum 2007 Jaarverslag transforum: duurzaam onder-nemen met kennis
Zoetermeer
W08 Wageningen ur 2008 Science for impact: on science, society and business
Wageningen
W09a WiaS 2009 documentation over 2003-2008 for Peer review of the graduate School WiaS
Wageningen
W09b WiaS 2009 report of the international Peer review of the graduate School WiaS
Wageningen
n10 nWo 2010 Begroting 2010 en meerjarencijfers 2011 tot en met 2014
den haag
chapter 7.
165
conclusions and discussion
7.1 Brief recapitulation
This thesis deals with the struggles for relevance of university researchers, their efforts to make their
work correspond with ruling standards of relevance, and to influence these standards. it was written
against the background of intensive debates about transformations in the knowledge infrastructure.
under labels such as ‘Mode 2 knowledge production’, ‘Post-academic science’, and ‘The entrepreneurial
university’, prominent scholars have claimed that university research is increasingly oriented to practical
applications in industry or policy, and that academic reward systems increasingly stimulate research that
is relevant to society and the economy rather than purely scholarly contributions (see chapter 2 for a
systematic review). although these diagnoses of the changing nature of university research have been
quite influential both in scholarly thinking and in policy circles, they are also surrounded by uncertainty
and confusion.
i have aimed to contribute to the understanding of transformations in the knowledge infrastruc-
ture by investigating struggles for relevance in three scientific disciplines in the netherlands. after the
introductory chapter, which explained this aim, specified the leading research questions, and introduced
the methods and the general theoretical framework, i started with ‘re-thinking new knowledge produc-
tion’ (chapter 2), a systematic analysis of the existing literature on changing science systems mentioned
above. This review yielded a number of problems related to the notion of Mode 2 knowledge produc-
tion and questions about its attributes transdisciplinarity, reflexivity and novel quality control. These
questions served to further sharpen my perspective and to choose priorities. in chapter 3 i developed
a preliminary theoretical framework consisting of the credibility cycle and a contract between science
and society, to guide my empirical studies. The following three chapters used and enriched (elements
of) this framework in the analysis of case studies of dutch chemistry, biology, and agricultural science.
in each case i addressed similar questions about changing struggles for relevance. chapter 4 reported on
the chemical case study and explored the different roles of practical applications in the credibility cycles
of catalysis, biochemistry and environmental chemistry. chapter 5 compared two biological fields,
paleo-ecology and toxicology, paying specific attention to growing multidisciplinary collaborations.
166
chapter 7
in chapter 6 three fields of agricultural science were analyzed, again focussing on the changing role of
stakeholder interactions in academic research practices.
This final chapter will provide a further reflection on the material presented in the previous chapters.
The main research question of this thesis is:
How to understand changes in the struggle for relevance of Dutch academic researchers in chemistry, biol-
ogy and agricultural science, in the period 1975-2005?
in chapter 1 i have specified three sub-questions:
1. What changes can be discerned in their struggles for relevance?
2. how can the similarities among the changes observed in different scientific fields be explained?
3. how can the differences among the changes observed in different scientific fields be explained?
These sub-questions will be subsequently addressed in the following three sections. at the end of Sec-
tion 7.4 i will provide an answer to the main research question. in Section 7.5 i will reflect on the main
concepts used in this thesis. Section 7.6 takes stock of the contributions this work makes to the literature
about changing science systems and presents some recommendations for further research. i will close
with a consideration of the policy implications of this study (7.7) and some concluding remarks (7.8).
7.2 Changes in the struggle for relevance
The starting assumption of this thesis was that academic researchers struggle for relevance: they try to
make their work correspond with ruling standards of relevance while simultaneously influencing these
standards (see Section 1.2). My first sub-question was what changes can be discerned between 1975 and
2005 in three scientific disciplines in the netherlands. in this section i will answer this question, based
on the empirical material presented in chapters 4-6. i have observed common trends at three steps in
the credibility cycle (see figure 7.1), at the acquisition of data, recognition and money, respectively.
167
conclusions and discussion
Figure 7.1 The three common trends in the struggles for relevance of eight fields of natural science, depicted in the credibi-lity cycle (adapted from latour and Woolar (1986)).
Intensification of struggle for relevance during data collectionfirst, in most fields the struggle for relevance during the collection of data (the actual research process)
has intensified. over the past few decades, the role of societal stakeholders in this process has grown.
nowadays, some researchers even collaborate so intensively with knowledge users that they conceive
them as ‘partners’:
‘We are free and they are free to go where they want. But you simply feel that they care about a long-lasting
relationship. So we don’t have a collaboration that is project-oriented. Then you would talk about custom-
ers. We talk about partners, [when we talk about] the firms.’ (researcher A7).
in agreement with the claims about the rise of ‘Mode 2 knowledge production’, an increasing share of
all dutch academic research is conducted in ‘the context of application’. in several fields interactions
with possible users of research outcomes were already common practice, but in general their frequency
and salience have grown. in fields like catalysis, aBg, and toxicology most projects are supervised by an
industrial sponsor or by a ‘users committee’. These receive a regular update about the progress and pro-
vide feedback for future directions. in other fields, like aPS and environmental chemistry, it has become
common to conduct academic research projects in collaboration with applied researchers employed
by public research institutes or private r&d labs. These types of interactions with stakeholders have
3. Intensi�cation of struggle for
relevance
Recognition
2. Decreasing value of practical applications
Data
Money
Articles (and other outcomes)
Sta and equipment
Arguments
1. Intensi�cation of struggle for relevance
168
chapter 7
increased the awareness of scientists of potential applications of their work and – to some extent - this
awareness influences their choices on the lab floor. There are also fields in which researchers still hardly
interact directly with societal stakeholders, in particular biochemistry and cell biology.
Decreasing value of practical applications as a source of recognitionThe second major development that was visible in all fields i have studied relates to the way scientists
earn recognition: on average, the value of practical applications as a source of academic recognition has
decreased. asked whether practical applications can help to get peer recognition, a biologist replied:
‘Maybe it would count a bit, but also here the scientific content comes first and if it is applicable by accident,
then that is so much to the good. I mean, it is like the cream on the pudding, but… It makes it more fun,
yes. But it is not…’ (researcher B20).
Since the 1970s recognition has become more and more linked to the production of scientific papers
(see text box 7.1 for illustrations). notably, this trend was hardly mentioned in the dominant ‘diagnoses’
of the changing science system (see chapter 2). Writings about academic capitalism, post-normal
science or Mode 2 knowledge production seem to neglect this salient development. Scientists are
increasingly under pressure to be produc-
tive, both in quantitative and qualitative
terms. in the credibility cycle, recognition
has become so strongly based on numerical
indicators of scientific productivity, that
academic researchers simply face the choice
‘publish or perish’. a strong publication list
has become a crucial condition in order to
qualify for particular kinds of funding, in
particular grants from national research
councils and from eu framework Program-
mes. Moreover, scientific productivity is the
main topic discussed during individual
performance interviews, and it is also the
main criterion for selecting candidates for
academic positions. in metaphorical terms, a
publication list needs length and depth. one
needs to write many papers, and besides it is
important that a certain share of one’s
‘in the past it was like… yes… you knew that
you were doing good stuff and it didn’t matter
so much whether you would have one or two
publications a year. nowadays that wouldn’t be a
lot.’ (researcher B18).
‘Yes, that is the most important principle: publi-
shing in the best possible journals.’ (researcher
c13).
‘first of all you want to make your findings pu-
blicly known. it is a way to get recognition from
your peers. Second it is also dire necessity, to be
able to give continuity to your funding. Because if
you don’t have publications… it is the way for the
outside world to assess your group.’ (researcher
c2).
Text box 7.1 interview quotes illustrating the increasing value of scientific publications as a source of recognition
169
conclusions and discussion
papers is of excellent quality. The latter is only partly evaluated by the content of a paper; it is usually
measured in terms of the impact-factor of the journal it is published in, or the number of citations it has
received. recently, a new indicator combining both aspects of productivity has gained prominence, the
so-called hirsch-index. in spite of its methodological limitations (costas & Bordons 2007, glänzel
2006), this indicator has become popular both in informal communication, and in official evaluation
processes134.
our interview data shows that scientists anticipate on the importance of scientific publications dur-
ing the whole credibility cycle. data collection is organized in such a way to optimize publication pros-
pects. Some scientists choose particular strategies because they either need to publish in a high-impact
journal (researcher B20) or because they perceive the need for a larger number of papers (researcher
a10), depending on the current status of their publication list. in addition, several ‘cheating’ strategies
are pursued, like the formation of writing ‘task-forces’ whose members grant each other co-authorships
without any actual collaboration (researcher c5), submitting several papers each highlighting different
aspects of the same research project, or dividing a particular contribution into its ‘smallest publishable
units’ (researcher a10).
over the years, publication achievements have (further) pushed away social or economic impact as a
source of academic recognition. as i will specify below, practical applications are not always in competi-
tion with scientific papers, but in most fields they are. in the selection of manuscripts for publication in
scientific journals, the societal relevance of the reported research does not play a significant role. obvi-
ously, editors and peers decide about to publish papers based on purely scientific considerations, like
consistency, novelty value and methodological quality. in many fields there is even a trade-off between
research projects that are of high societal relevance and projects that are likely to result in (many) high-
impact scientific papers.
Intensification of struggle for relevance in context of funding acquisitionThe third generic trend that was visible in the period 1975-2005 concerns the intensification of the
struggle for relevance in the context of funding acquisition. My findings confirm the claims in studies
about academic capitalism (Slaughter & leslie 1997)(also see Section 2.3.5 of this thesis) that funding
acquisition has become an increasingly competitive affair. earning sufficient income for continuing one’s
research activities is not anymore self-evident, but it has become the result of active acquisition efforts.
The process of acquiring money for continuing and expanding research activities has become laborious
and time-consuming. My interview data raise the impression that it has become common for senior re-
134 for example, in the documentation prepared for the most recent evaluation of the Wageningen institute of animal Sci-ences, the hirsch-index of every scientist was listed (WiaS, 2009, ‘documentation over 2003-2008 for Peer review of graduate School WiaS’, Wageningen).
170
chapter 7
searchers to spend between 10 and 20 % of their time on networking, exploring funding options, writing
proposals and negotiating contracts.
Promises about practical applications often play a central role in the selection of project proposals.
in other words: the meaning of relevance has changed. in all three disciplines i have studied, i could
identify changes in the meaning of relevance (see table 7.1). in chemistry and biology, in the 1970s rel-
evance was mainly expressed in terms of the education of new generations of researchers, but later it was
increasingly defined as producing strategic knowledge that can be used to improve industrial products
and processes, to create medical innovations or to design public policy. in addition, nowadays academic
involvement in the valorization of knowledge by creating spin-off firms, exploiting patents or consul-
tancy services is also regarded as a sign of relevance. in the case of dutch agricultural science practical
applications have always been central in the notion of relevance. however, the content of this notion has
shifted from contributing to productivity enhancement to contributing to sustainable farming.
Table 7.1 Shifts in the meaning of relevance in three scientific disciplines
chemistry Biology agricultural science
Main conception of relevance ~ 1975
education education Productivity enhancement of agricultural production
Main conception of relevance ~ 2005
industrial innovationsvalorization
industrial innovationsMedical applicationsenvironmental policyvalorization
More sustainable agricultural production
in all fields the scientists i have interviewed report that aligning their work with the knowledge needs
of societal stakeholders has become increasingly important for gathering sufficient funding. in the
credibility cycle, expected societal benefits strongly catalyze the conversion of recognition into money.
Based on a certain amount of recognition (for example, expressed in one’s publication list), the same
researcher will more readily acquire funding if (s)he manages to convincingly specify the societal value
of a proposed research project:
‘Yes. It is much easier to get money, there are much more possible sources to get money, if you have some-
thing that is relevant to society.’ (researcher C12)
Still there are some possibilities to get funding for research without promising societal benefits, most
at the national research council nWo. But for only a few of the 47 researchers i have interviewed these
provide a substantial share of their budget.
The three trends just described are not necessarily related; the latter two even seem to point in opposite
directions. in general, funding acquisition has become increasingly dependent on expected societal ben-
171
conclusions and discussion
efits, while peer recognition has become increasingly dependent on academic achievements. however,
no general observations can be made about the interplay of these two developments, as the relationship
between practical applications and scientific productivity turns out to vary strongly across fields. in the
following the differences among the changing struggles for relevance in different scientific fields will be
presented, including a further exploration of the possible tension between practical applications and
scientific productivity.
Changing struggles in different scientific fieldsto a certain extent the three common trends just presented were visible in all eight fields (in the
netherlands). however, a closer look also reveals significant differences across scientific fields in the
manifestation of these changes. Surprisingly, the differences among fields within each discipline i have
studied turn out even more significant than the differences among the aggregated disciplines themselves.
Some interesting conclusions can be drawn from the comparison of chemistry, biology and agricultural
sciences. But, thanks to the diversity observed within each discipline, the most powerful insights can be
gained by comparing individual fields.
in my findings, four types differences can be discerned among the eight fields studied, regarding the
changes in their struggles for relevance (see table 7.2). The first three are directly linked to the general
trends just described. a fourth concerns the tension that has arisen in some fields due to the combina-
tion of these trends. first, although the role of stakeholders in the academic research process has general-
ly grown, the extent to which stakeholders have become involved in data collection varies strongly across
fields. Second, there is variation in the (limited) degree to which practical relevance is rewarded in terms
of academic recognition. Third, in all fields promising societal benefits can help to acquire funding, but
the degree of involvement of societal stakeholders in the actual agenda-setting differs. a fourth dimen-
sion that deserves to be addressed here is the relationship between practical applications and scientific
publications; in some fields there is a synergy between scientific productivity and practical relevance; in
other fields these are in contradiction.
Table 7.2 overview of the differences among struggles for relevance in eight scientific fields
Credibility conversion Variable Observed range
acquiring data intensity of stakeholder interactions low – high
acquiring recognition The value of practical applications negligible – considerable
acquiring money influence of stakeholders on research agenda Weak – Strong
(generic) relationship practical applications – scienti-fic productivity
Strong tension - Synergy
172
chapter 7
ChemistryBecause chemistry is traditionally committed both to a quest for fundamental understanding and to con-
siderations of use, in Section 1.5 i expressed the expectation that it could relatively easily accommodate
the changes in its institutional environment. given their tradition of ‘use-inspired basic research’ (Stokes
1997), it was assumed that chemical researchers would not have difficulties with the increasing pressure
for publications and the practical orientation of many funding sources. on average, my findings confirm
this expectation, but a closer look reveals considerable differences across the three fields.
in dutch catalysis i observed a further increase in the intensity of interactions with chemical indus-
try. The knowledge needs of chemical firms play a significant role in most types of funding. Moreover, in
a large share of all projects, firms participate in a users committee or supervisory board that has a say in
the progress and direction of data collection. academic researchers in catalysis are often involved in pat-
ent applications and in high tech startup companies. in addition, their scientific publications and confer-
ence presentations also serve as transfer channels to industry, as industrial representatives (or industrial
researchers) read scientific journals and attend scientific conferences. The pressure for publications does
not inhibit interactions with industry, as application-oriented research is very suitable for academic
publications. contributions to practical applications are not rewarded in terms of scientific recognition,
career opportunities, or evaluation scores, but they do not seem to hinder these rewards either. overall,
in this field the relationship between excellent performance in academic terms and producing relevant
knowledge is synergetic rather than counterproductive.
in biochemistry interactions with stakeholders have not increased significantly since the 1970s.
The context of application is still hardly present in this field. in order to enhance their chances to get
funding, biochemical scientists have started to loosely refer to possible practical applications in research
proposals, but they do not typically interact with stakeholders to set their research agenda. Most funding
can be acquired based on rather general promises about future applications in health care or biotechnol-
ogy. The data collection is carried out autonomously, without (frequent) interactions with stakeholders.
dutch biochemists are also not involved in knowledge transfer to industry or other end-users, they only
collaborate with applied researchers, e.g. in the medical domain. to earn recognition, to receive high
evaluation scores and to develop an academic career, it has become extremely important to publish
many papers, to be cited and to acquire research grants. overall, the increased pressure to publish and
the scarcity of resources have further increased the competitiveness of this field, but the struggle for
relevance has remained a side issue.
in dutch environmental chemistry, the struggle for relevance has remained rather stable over the
past few decades. interactions with stakeholders have been common practice in this field, since its incep-
tion in the 1960s. Both during funding acquisition and data collection environmental chemists attune
their activities to the needs of their users, i.e. policy-makers, firms, or ngos. They also actively transfer
their findings to users by writing policy reports, participating in advisory committees, and contributing
173
conclusions and discussion
to public debate. These aspects of the struggle for relevance have not changed a lot over the past few
decades. however, the increased pressure for academic publications has complicated this struggle. The
intensive interactions with stakeholders carry two dangers: they can threaten the objective position of
academic researchers and they can erode the available time for writing scientific papers. in this field,
(more than in most other fields), practical applications can help to receive recognition from academic
colleagues. however, in formal quality control (performance evaluations, performance interviews, job
interviews) only scientific publications count as real success.
Biologyin biology i expected to find relatively weak (existing) relationships with societal stakeholders. This
would imply that it could be difficult to accommodate the increasing pressure for practical applications.
however, in one of the two biological fields (toxicology) strong relationships with knowledge users
already existed, while in the other (paleo-ecology) they have developed recently without any problems.
in fact, the enduringly weak relationships with stakeholders expected in dutch biology was more closely
approximated in biochemistry and cell biology, two fields with close affinity with biology, but which
were part of my case studies of chemistry and agricultural science, respectively.
in dutch paleo-ecology the concern with practical applications has increased in the period studied.
traditionally this field had a quite academic orientation, directed at enhancing the understanding of
taxonomy and evolution, but in the 1980s collaborations with oil companies were started in order to
ensure sufficient funding. in addition, nowadays a major share of paleo-ecologists is involved in climate
research. in this type of work, agenda-setting is strongly connected with policy needs. This new research
direction does not require active involvement of knowledge users in the actual research process, but the
collaboration with other disciplines (such as earth scientists) is crucial. By this type of multidisciplinary
research, paleo-ecologists can provide direct or indirect support for environmental policy. like in most
other fields, career advancement, performance evaluations and informal reputations are based much
more strongly on scientific productivity than on policy relevance. But the increasing publication pres-
sure does not conflict with the increasing pressure for relevance. Paleo-ecologists seem able to produce
prestigious papers by participating in climate-oriented programs or by contract research for industry just
as well as by projects with a more fundamental orientation.
The dynamics of dutch toxicology are similar to those of environmental chemistry, in the sense that
the struggle for relevance has not intensified in these fields. academic toxicology has been connected
with practical applications since its emergence in the 1970s. interactions with knowledge users (such as
policy makers) have always played a significant role in funding acquisition, data collection and com-
munication in this field. over the years, toxicologists have experienced a growing pressure to perform
better according to academic standards. due to changing knowledge demands and a growing publishing
pressure, a more fundamental research orientation has developed focusing on the mechanisms behind
174
chapter 7
toxicity. Still, a major share of the available funding is connected to practical goals such as exploring the
medical risks of a particular class of compounds. But within projects of this kind, the role of academic
toxicologists is to produce generic models and understanding rather than performing standard toxicity
tests. Knowledge exchange with users such as policy makers, firms, and medical scientists is still com-
mon practice, but efforts in this direction are less rewarding in terms of career advancement and peer
recognition than producing papers for scientific journals.
Agricultural sciencein agreement with my expectations (expressed in Section 1.5), in agricultural science the increased pres-
sure for scientific publications of the past few decades has had the strongest impact. as this discipline
was traditionally relatively weakly oriented to fundamental understanding, the demand for scientific
productivity was more disrupting here than in chemistry and biology. But the effects of this develop-
ment also varied across the three fields. in animal breeding and genetics (aBg), similar to catalysis,
researchers had no problems with increasing their publication output while at the same time producing
knowledge of strategic importance to their stakeholders. in animal production systems aPS and cell bio-
logy, however, many researchers face a trade-off between activities that can help to produce prestigious
publications and activities promising high practical utility for their stakeholders.
in aBg the interactions with stakeholders have further increased over the past 30 years. Similar to
catalysis, this field has a strong relationship with a specific industrial sector, animal breeding firms. The
university researchers have intensive contact with these firms in all stages of their work. Some of the
projects are funded by industry; others are (mainly) funded from public sources. But in all cases, the
long term needs of breeding firms are taken into account in the choice of research goals. in many cases
animal breeding researchers also interact with firms during data collection, as these firms provide data or
participate in users committees or similar bodies supervising the projects. aBg researchers intensively
collaborate with the applied research institute Wageningen ur livestock research. This institute also
mediates some of the relationships with industry. The outcomes of aBg’s research seem to reach poten-
tial users easily, thanks to the good contacts, and thanks to the high absorptive capacity of the breeding
firms. These companies are highly knowledge intensive and they employ a lot of aBg’s graduates. aBg
researchers are not particularly rewarded for (contributions to) practical applications of their knowledge
in terms of career advancement or peer recognition. Still these contributions seem to catalyze rather
than hamper academic publications. industrial support facilitates research oriented to fundamental
understanding, the outcomes of which are suitable for publication in prestigious scientific journals.
aPS is a relatively young field, which has developed strong relationships with societal stakeholders.
its funding comes from a variety of sources, but it is most often linked to promised practical applications.
aPS’s research agenda is strongly connected to the needs of stakeholders, although these stakeholders
are quite diverse. farmers, the end-users of this knowledge, do not provide funding, but they do contrib-
175
conclusions and discussion
ute to the collection of data. The outcomes of aPS research are not only published in scientific papers
but also delivered to possible users, in the netherlands or abroad. The increased pressure to publish is
a challenge for aPS. The interdisciplinary and application-oriented nature of its research makes it more
difficult to publish in prestigious journals. in addition, due to the novelty of the aPS approach, there is
a limited disciplinary infrastructure available, in terms of specialized journals and research communi-
ties. in this field practical applications can help to earn informal recognition from colleagues, as they
acknowledge them as legitimate outcomes. however, applications hardly help to receive high evaluation
scores or to enhance career prospects.
Within dutch agricultural science, cell biology has always been a relatively fundamental field and in
the past 30 years it has not increased its interactions with societal stakeholders a lot. The research agenda
is still driven by academic concerns. Stakeholders, such as firms, provide only a marginal share of the
research funding. also the processes of data collection and publication are carried out without frequent
stakeholder contact. The increasing competition for research council funding and the increasing pressure
to publish stimulate cB to shift towards more ‘prestigious’ research areas that are of less relevance for
agricultural practice. More money is available for research on mouse or human models systems, and this
also fits better in high-impact journals. This situation creates a difference between research that is most
useful for agricultural stakeholders and research that provides the best prospects on academic rewards
such as publications, research grants, evaluation scores and career advancement.
The relationship between practical applications and scientific productivityhaving presented my findings regarding the eight fields included in my case studies, let me now make
some observations cutting across disciplinary boundaries. When comparing the changing struggles for
relevance of the eight fields studied, a particularly interesting dimension is the relationship between
practical applications (which have become dominant in the meaning of ‘relevance’) and scientific
productivity. as the findings reported above indicate, in some fields there is a tension between practical
applications and scientific productivity, while in other fields they have a synergetic relation. table 7.3
presents a classification of the eight fields in my sample based on this dimension. in the following some
the characteristics of the three categories of fields will be explored in terms of the other three ‘dimen-
sions’ of the struggle for relevance.
176
chapter 7
Table 7.3 classification of the eight fields based on the relationship between practical applications and scientific producti-vity
Relationshippractical applica-tions - scientific productivity
Fields
Intensity of stake-holder interac-tions during data collection
Value of practi-cal applications for acquiring recognition
Influence of stakeholders on research agenda
Strong tensioncell Biology Slight increase negligible remains weak
Biochemistry Still low negligible remains weak
Weak tension
environmental chemistry high but stable considerable Strong but
stable
animal Produc-tion Systems high and growing considerable Strong and
increasing
toxicology high but stable low Strong, slight increase
Synergy
catalysis high and growing negligible Strong and increasing
animal Breeding & genetics high and growing low Strong and
increasing
Paleo-ecology Slight increase low increasing
in the two fields with the least intensive stakeholder interactions, biochemistry and cell biology, i ob-
served a trade-off between scientific productivity and practical relevance. in these fields scientists com-
plain that engaging in application-oriented research projects and interacting with societal stakeholders
‘distracts’ them from the main focus of their field. efforts or achievements of this kind do not significant-
ly yield peer recognition here. in biochemistry and cell biology contributions to high-impact journals are
usually based on projects paid by research councils granting considerable autonomy to the researchers to
formulate their own research priorities and approaches. enhancing the industrial, medical or agricultural
relevance of one’s work implies a move away from the central debates of these fields. to this end, one
would have to shift to other model systems (for example chicken rather than mouse) or to address other
research questions (for example relating to specific treatments rather than general understanding) which
are less suitable for publishing in prestigious journals.
in a second class of fields, containing aPS, environmental chemistry and toxicology, such a ten-
sion exists as well, but it is of a weaker nature. in this set of fields interactions with stakeholders are
quite common, both during the acquisition of funding and during data collection. here, under certain
conditions, application-oriented research can lead to impressive publications. The most important
requirements for successfully combining practical applications with scientific productivity seem to be
substantial project size and consistency across projects. if these conditions are fulfilled, the results of
177
conclusions and discussion
application-oriented projects can lead to improved understanding on a fundamental level. in this way,
the outcomes of (one or more) relatively practical projects can lead to scientific papers. for researchers
involved in relatively short and diverse application-oriented projects it is difficult to develop fundamen-
tal insights that can be published in prestigious scientific journals.
The situation is different in catalysis, paleo-ecology and aBg. in this class of fields, in which
stakeholder interactions have significantly grown, i observed a synergy rather than a trade-off between
scientific productivity and societal relevance. interactions with stakeholders do not only help to acquire
funding, but they are also helpful in other credibility conversions. i identified three mechanisms that are
responsible for this synergy. first, applied research projects for stakeholders can provide access to data
that are useful for more fundamental investigations, too. Second, the interactions with stakeholders often
give inspiration for challenging research questions. Third, some stakeholders simply sponsor fundamen-
tal research activities, of which they expect benefits on the longer run.
as table 7.3 indicates, the relative value of applications (when compared to publications) as a source
of recognition does not show a clear trend across the three sets of fields, that is, it does not correlate
with the degree of synergy between practical applications and scientific productivity. it is the lowest in
biochemistry and cell biology, the fields with the least intensive stakeholder interactions. in these fields,
recognition is almost exclusively based on academic achievements, in terms of scientific publications and
citations. in the two other classes of fields the relative value of applications ranges from negligible to con-
siderable. Practical applications are most rewarding in terms of recognition in aPS and environmental
chemistry, which occupy a middle position in table 7.3. in these two fields a scientific reputation is not
only based on contributions to scientific debate, but also to contributions to environmental policy or to
the development of more sustainable agriculture. This may be related to the fact that in these two fields
scientists seem most strongly motivated to ‘change the world’. More than in the other fields studied, they
draw inspiration for their work from personal ambitions to contribute to external goals like sustainable
development135. in the same vein, they also value their colleagues’ practical contributions to such goals,
more than scientists in fields like catalysis or toxicology.
Sub-question 1to conclude this section, i have observed three general changes in the struggles for relevance of dutch
chemistry, biology and agricultural science, in the period 1975-2005:
1. The struggle for relevance during data collection has intensified.
2. The value of practical applications as a source of academic recognition has decreased.
3. The struggle for relevance in the context of funding acquisition has intensified.
135 for example: ‘Yes, in the time that i started i had a strong passion. That there was a large problem which already received attention, but which was not known yet in its full proportions.’ (interview researcher c15).
178
chapter 7
Moreover i have discerned significant differences in the way these changes became manifest in differ-
ent scientific fields, and in the interplay between them. in some fields a tension has developed between
scientific productivity and practical applications, in others a synergy.
7.3 Explaining the similarities among scientific fields
The work reported in this thesis shows that the three similarities among the changing struggles for rele-
vance of the different fields can be understood as effects of the combination of three structural changes:
shifts in the available funding, the rise of performance evaluations, and changing views on the societal
position of the university. chapters 4-6 presented an analysis of the changing structures influencing the
work of academic researchers in chemistry, biology and agricultural science. using the science-society
contract as an entrance point, in chapters 4 and 5 the main emphasis was on the discursive dimension
of structure. chapter 6 explored in more detail the institutional dimensions, based on the notion of a re-
search system. from both angles, i have gained an understanding of the way structural conditions shape
the agency of scientific researchers. The three structural changes i observed in all cases together provide
a possible explanation for the generic changes i have observed in the credibility cycle: the intensifying
struggle for relevance during data collection, the decreasing value of relevance as a source of recognition,
and the intensifying struggle for relevance in funding acquisition (see figure 7.2). Below i will summa-
rize my findings regarding the three structural changes in the different case studies, complemented with
some generic data where possible.
Figure 7.2 overview of the effects of structural changes on the struggle for relevance
Diversification of fundingThe first general structural change influencing struggles for relevance in all fields of science, is trend of
funding diversification. in all fields studied the relative share of public funding for basic research has
Diversi�cation of funding
Rise of performance evaluations
Changing views on societal position of universities
Intensifying struggle for relevanceduring data collection
Decreasing value of relevance as a source of recognition
Intensifying struggle for relevance during funding acquisition
Diversi�cation of fundff ing
RiRR se of perfoff rmanceevaluations
Changing views on societalposition of universities
Intensifyff ing struggle foff r relevanceduring data collection
Decreasing value of relevance as asource of recognition
Intensifyff ing struggle foff r relevanceduring fundff ing acquisition
179
conclusions and discussion
decreased. Moreover, over the years the relative share of unconditional funding (first money stream) has
decreased. With the general expansion the public science systems, budgets have become under pressure
and the need to account for public investments in academic research has grown (Ziman 1994). in line
with the ideologies of neo-liberalism and ‘new Public Management’, the government has loosened state
control and introduced market mechanisms to enhance efficiency and effectiveness (de Boer et al. 2007,
Schmoch & Schubert forthcoming). Since 1975, the starting point of my analysis, the government has
transferred an increasing share of public funding to competitive arrangements, organized by research
councils or other intermediary organizations. Between 1975 and 2005 the total amount of block grant
support for universities has grown almost twofold in real terms (versleijen 2007), but its relative share
in relation to more competitive funding sources has decreased. in 1985 governmental block grants still
constituted 84% of all revenues of dutch universities, but in 2000 this was only 70%. in relation to the
number of enrolled students, the block grant support has even decreased with 30% between 1980 and
2000136 ( Jongbloed & Salerno 2003). around 1975, this money stream was still sufficient for research
groups to buy the necessary equipment and hire some temporary staff, but nowadays even some of the
permanent academic staff needs to be paid from project funding. even the funding that has remained in
this category has become less secure, as it has become subject to university policy, and it is often needed
to ‘match’ externally acquired funding ( Jongbloed & Salerno 2003, aWt 2004).
The ‘second money stream’ has also changed dramatically. research councils were initially organized
in sub-disciplinary Working committees, but they have been merged and reorganized into a general
matrix organization supplying most funding in the form of multidisciplinary research programs. While
dutch research councils originally exclusively funded basic research, they have expanded their territories
to application-oriented activities, too. Moreover the dutch organization for Scientific research (nWo,
the new umbrella organization of all research councils) has developed a variety of hybrid funding con-
figurations in collaboration with ministries, firms or other knowledge users.
in addition, the third money stream (all contract funding except from nWo), which is more
strongly oriented to practical applications, shows a spectacular increase in the 1980s and 1990s at all
dutch universities (see figure 7.3). Between 1983 and 2000 the total size of this ‘stream’ has increased
from about 125 to about 638 Million euros. This amounts to an increase of a factor 3.85 in real terms
( Jongbloed & Salerno 2003).
136 at Wageningen university, one of the few dutch universities with decreasing student numbers, governmental block grant support even decreased with about 56% in relation to the number of students during the 1980s (in real terms (faber 1993), see also chapter 6). Between 1978 and 1995, the number of research positions paid from block grant funding at this university increased from 223 to 361, while those paid from external funding increased 77 to 472 (rose-boom & rutten 1998).
180
chapter 7
Figure 7.3 The size of the ‘third money stream’ in 1983 and 2000 (in current prices) at dutch universities. Source: Jong-bloed & Salerno, 2003.
in the case of chemistry the willingness of industry to support academic research has grown a lot. due
to the adoption of new r&d strategies involving less in-house basic research, firms have become more
dependent on knowledge produced outside their own laboratories, so they provide a substantial share
of all funding for academic chemistry. Biology has a less straightforward relationship with one particular
industrial sector, but it also receives increasing support from third parties, such as governments, ngos,
medical charities, and the european framework Programmes. in the case of agricultural science, bio-
technology firms, ngos and new intermediary organizations have taken positions as significant funding
sources next to the traditionally dominant Ministry of agriculture.
These shifts help to explain the intensification of the struggle for relevance during data collection and
during funding acquisition. in all three disciplines a strong diversification of funding sources has taken
place between 1975 and 2005 and the relative share of unconditional support granting complete auton-
omy has decreased. overall the funding system has become more competitive. academic recognition as
such is not anymore a guarantee for continuous funding. nowadays the large majority of projects is paid
from external funding, based on carefully written research proposals and / or negotiations. and within
the palette of available additional funding, the share of money tied to specific practical or strategic goals
has increased from a marginal element to the dominant majority. in this situation, academic researchers
need to maintain relationships with all types of organizations that may serve as future funding sources.
Simultaneous to this major shift towards application-oriented funding, however, a smaller trend took
place in the opposite direction. over the last decade, the research council’s policy to nurture and stimu-
late excellent researchers has created a small but significant subset of funding arrangements lacking any
consideration of practical utility. in 2000 nWo introduced the highly competitive ‘vernieuwingsimpuls’
grants, as a policy-instrument for supporting talented researchers. in the selection of proposals for these
181
conclusions and discussion
grants, the most determining factor is the individual quality of the requesting applicant, assessed mainly
using bibliometric criteria. With its emphasis on bibliometric quality indicators, this type of funding has
probably contributed to the decreasing value of practical applications as a source of recognition. at the
moment the grants under this scheme together consume about 20% of nWo’s total budget137. however,
the relative impact of this funding instrument is probably larger than its financial share, thanks to its
prestige and popularity. Because these grants do not provide any thematic restrictions and they offer a
large degree of autonomy during the research process, they are generally considered as one of the most
attractive funding sources available. Between 2000 and 2006 the average number of applications for
the vernieuwingsimpuls was about five times the number of available grants (technopolis & dialogic
2007). This implies that the success rate of this funding instrument is only 20%, while the overall aver-
age success rate at nWo is about 50%138. Moreover, because of their competitive nature, successful
acquisition of vernieuwingsimpuls-grants counts as a strong quality indicator both in formal evaluations
(performance interviews, job interviews, research evaluations) and in informal reputation. for these
reasons, the stimulating effect of this funding scheme reaches a much larger population of researchers
than the grant-winners only.
Rise of performance evaluationsThe second structural change that has shaped struggles for relevance in all scientific fields studied, is the
rise of performance evaluations. This development has contributed to the increasing pressure for scien-
tific publications and the decreasing value of practical applications as a source of academic recognition.
over the past 30 years society’s general need for accountability has led to the development of an elabo-
rate system of performance evaluations of all academic research. inspired by the ideology of new Public
Management, the government has gradually developed an approach of ‘steering at a distance’, granting
more autonomy to universities as organizational units (de Boer et al. 2007). This new governance
model implies the need for more management information, such as numerical indicators about the
performance of specific organizational units (Schmoch & Schubert forthcoming). research evaluations
could meet this need. after a number of pilot evaluations and foresight studies in the 1980s, a more
or less standard approach has been developed for systematic evaluations of academic research groups
(van der Meulen 2008). This trend can be seen as a manifestation of a wider ‘audit explosion’ in the late
1980s, in response to increasing demands for accountability and transparency (Power 1999). nowadays,
every research group in the netherlands is subject to regular evaluations. research quality assessments
officially use a variety of criteria139, but in practice they tend to be ruled by bibliometric quality indica-
tors. even if other dimensions such as viability or relevance are measured as well, in the interpretation of
137 nWo, 2010, ‘Begroting 2010 en meerjarencijfers 2011 tot en met 2014’, nWo, den haag138 This was the average success rate in 2004 (nWo, 2005, ‘Jaarboek 2004’, nWo, den haag)139 vSnu, nWo & KnaW 2003, Standard Evaluation Protocol 2003-2009 for Public Research Organisations
182
chapter 7
the evaluation scores, numbers of publications and citation rates dominate140. The availability of digital
bibliometric databases, and the relative generic validity and cross-comparability of these indicators made
these into a success that is not equalled (yet) by any other indicators (gläser & laudel 2007). unlike
the research assessment exercise in the uK (Barker 2007) the results of dutch research evaluations do
not have direct financial consequences. however, the outcomes do influence strategic decisions by deans
and university boards, and high scores can also contribute to successful acquisition of external funding.
Changing views on the societal position of the universityThe third important structural change concerns the dominant perspective on the mission of the univer-
sity in the netherlands. in the period between WWii and the start of science policy in 1975, the ruling
values connected with academic research were autonomy and independence. in the next 30 years, these
values have been gradually replaced by accountability, valorization and entrepreneurship.
These shifting views can be traced in the rationales for funding academic research. in the rationales
for funding academic chemistry and biology, the emphasis has shifted from education and cultural
value to practical applications. Before 1975 chemical and biological research was mainly considered
valuable because it facilitated the training of new generations of researchers. as a minor argument one
sometimes referred to the intrinsic (cultural) value of basic research. Since the 1970s public support for
these disciplines has increasingly been justified with reference to their contributions to the solution of
societal problems and their contribution to industrial innovations. over the course of the years, society
has developed different expectations of the contribution academic scientists should make to social and
economic goals, giving them an increasingly active role in reaching these goals. in the early days of sci-
ence policy, in the late 1970s, the rise of environmental concerns and critical views on scientific progress
created a demand for chemical expertise to contribute to the solution of environmental problems and
the development of cleaner technologies. later, in the 1980s, when innovation became an explicit policy
goal and when firms started to reorganize their r&d activities, funding for academic chemistry became
connected to the expectation that its outcomes provide starting points for chemical industry to develop
new products and processes. in the case of biology, the legitimacy for supporting academic toxicology
was already connected to environmental problems since the emergence of this field in the early 1970s.
Support for paleo-ecology became connected to economic rationales in the 1980s, when it started to
contribute to the identification of potential oil drilling locations. The commercial success of biotechno-
logical industry has also provided additional rationales for supporting academic biology, but these are
of little significance for the two fields focused on in chapter 5 (toxicology and paleo-ecology). Since the
140 The newest protocol for dutch research evaluations more explicitly demands the assessment of ‘societal relevance’ (vSnu, KnaW & nWo 2009, Standard Evaluation Protocol 2009-2015: Protocol for Research Assessment in the Nether-lands), and a recent set of pilot-studies has shown the potential of indicators for this criterion (eric 2010, Handreiking Evaluatie van maatschappelijke relevantie van wetenschappelijk onderzoek, eric publicatie 1001.), but the effects of this development on academic research practices were not yet visible in my case studies.
183
conclusions and discussion
early 1990s, the need for sustainable development is mentioned a lot in arguments for funding research
in both disciplines. This umbrella term covers potential contributions to both environmental issues and
economic prosperity. These contributions include enhanced understanding of climate change (by paleo-
ecology), higher efficiency of chemical production (by catalysis) or lower uncertainty about the risks of
synthetic compounds (by environmental chemistry). recently, an additional set of rationales for sup-
porting academic science has emerged under the label of ‘valorization’, connected to the potential direct
economic benefits biologists and chemists can generate. Patents and spin-off creation have become signs
of successful valorization, providing access to future funding from innovation-oriented policy instru-
ments (Zomer et al. 2010).
a similar development is visible in agricultural science, although with a different point of refer-
ence, as funding in this area has always been connected with expected practical applications. however,
also in this case a significant shift can be observed in the specific applications demanded and in the
visions on the optimal organization of the knowledge system. in the first post-war decades, support for
academic agricultural science was based on the widely shared need to enhance agricultural productiv-
ity by modernization. agricultural researchers could count on unconditional support, based on the
linear-model-based belief that the basic insights they generated would be unproblematically be turned
into technological innovations by applied researchers and diffused to farmers by the celebrated govern-
mental extension service. after dutch agricultural production reached its geographical, economic, and
environmental limits and undesirable side effects of agricultural modernization became visible, views
on the future of this sector radically diverged. although the level of consensus of the 1950s and 1960s
has never returned, nowadays the dominant perspective is the need for a (more) sustainable agriculture,
which is internationally competitive, without comprising the environment, animal welfare, or human
health. Moreover, the dominant views on an effective agricultural knowledge system have changed from
a consensus-driven homogeneous structure to a loosely coupled heterogeneous network representing
a variety of stakes. in the former, academic researchers had a clear role as the producers of fundamental
insights in agricultural production, and did not have to be directly involved in the transformation and
translation of these insights into actual farming practices. in the latter system, however, their role is much
larger and more diffuse than only knowledge production and transfer to applied researchers and exten-
sion agents. They are expected to interact with all actors in the agricultural ‘innovation system’, including
farmers, supplying firms, consumer organizations, environmental ngos, and regulating bodies. What is
more, these interactions should go beyond one-way communication as agricultural scientists are invited
to participate in ‘co-production’ and ‘transdisciplinary learning’ (hoes et al. 2008, regeer 2010).
The changing views on the societal position of the university have two important implications.
first, while the values autonomy and independence encourage scientists to isolate themselves from the
outside world, values such as valorization and entrepreneurship invite them to collaborate, participate
and interact. This has contributed to the intensification of struggles for relevance during the collection
184
chapter 7
of data. Second, the rise of accountability as a value implies that scientists need to show that they deliver
value for money, both in terms of what they are doing and how much they produce. This need has been
fulfilled with performance evaluations, which have changed the norms guiding peer recognition, as i
have explained above.
Structures are both medium and outcomePlease note that these explanations do not imply a unidirectional causality between the structural con-
ditions and the agency of university researchers. increasing interactions with societal stakeholders can
also lead to changes in the dominant view on the societal mission of universities, and to modifications in
the available funding for academic research. Paleo-ecology, for example, was traditionally seen as rather
fundamental, but got a more strategic image due to the development of stronger connections with oil
companies. This created new funding possibilities, both from public and from private sources.
Moreover, changing publication patterns influence the way performance evaluations are organized
and also the quality criteria used by research councils to distribute their funding. Bibliometric criteria
would never have become so dominant in the allocation of personal grants in the context of nWo’s
vernieuwingsimpuls if journal papers had not developed into the most universally accepted research
outcome. The declining prominence of dutch publications in fields like environmental chemistry and
toxicology is not only an effect of the rise of bibliometric evaluation criteria, but it also strengthens their
validity and legitimacy. it is probably appropriate to speak of a co-evolution of publication practices and
evaluation structures. due to limitations in time and space, feedbacks from agency to structures have
received less attention in this thesis, but i do not want to suggest that they are not significant.
Sub-question 2The conclusion of this section is that a possible explanation for the similarities among the changes in the
struggles for relevance in different scientific fields can be found in changes in the structural conditions of
academic research, in particular the diversification of funding, the rise of performance evaluations, and
changing views on the societal position of the university.
7.4 Explaining the differences among scientific fields
a notable result of this thesis is that struggles for relevance differ more strongly among fields than among
aggregated disciplines. While my case studies were selected on the level of disciplines, the comparison of
their constituent yields more interesting outcomes than the comparison of whole disciplines. in all three
disciplines the changes in the struggle for relevance varied considerably across fields. for example, inter-
actions with stakeholders have increased in both catalysis and environmental chemistry, but in catalysis
185
conclusions and discussion
these help scientists to build excellent publications record, while in environmental chemistry these tend
to constrain prestigious publications. it seems that the struggle for relevance of catalysis shares more
characteristics with the one in animal breeding & genetics than with biochemistry or environmental
chemistry. how can we explain the differences among the changing struggles for relevance of scientific
fields? This section will present a possible explanation based on socio-organizational, cognitive and
cultural field-characteristics, combined with the characteristics of societal stakeholders.
Explaining variation in stakeholder interactionsin the previous section we have seen that the degree to which stakeholder interactions have increased
(both during data collection and during funding acquisition) varies strongly across fields. This can be
partly explained by cognitive and socio-organizational characteristics of scientific fields, in particular
their search pattern (Bonaccorsi 2008) and strategic task uncertainty (Whitley 2000). i have found that
some fields with ‘convergent’ search patterns, namely biochemistry and cell biology, have developed
relatively few interactions with societal stakeholders (see tables 7.3 and 7.4). This is understandable,
because convergent fields have a relatively sharp focus in terms of research problems and approaches.
in these fields the strategic task uncertainty is low. This implies that there is a strong overall consensus
about the intellectual priorities and scientists can not easily develop a new, application-oriented research
direction.
Table 7.4 classification of fields based on the degree of convergence of their search pattern
Search pattern Chemistry Biology Agricultural science
convergent Biochemistrycatalysis toxicology cell biology
animal breeding and genetics
divergent environmental Paleo-ecology animal production systems
in divergent fields, such as aPS an environmental chemistry, it may be more likely that niches develop
which fit the knowledge needs of societal stakeholders. divergent fields typically have a high strategic
task uncertainty. This means that a large diversity of research directions is accepted simultaneously be-
cause there is no overall consensus about the intellectual priorities. researchers and employers are able
to pursue distinct strategies and orientations without being penalized for theoretical deviance. in such
fields it is easier to develop new research directions that fit the needs of societal stakeholders.
Explaining variation in the value of practical applications as a source of credibilityThe extent to which practical applications count as a source of credibility, another dimension of chang-
ing struggles for relevance, can be understood when taking into account the traditional communication
culture of a scientific field. fields with a divergent search pattern generally have a ‘rural’ communication
186
chapter 7
style (Becher & trowler 2001), as they are not very competitive and have a low people-to-problems
ratio. in rural fields, such as aPS an environmental chemistry, relatively few researchers work on a large
number of dispersed problems, and the mutual competition is limited. as there is no broad consensus
about overall quality standards, it is also difficult to formulate general evaluation criteria. due to the
theoretical diversity knowledge accumulation is less efficient than in ‘urban’ fields and there is usually a
lower citation density. We can assume that in these fields less high-impact journals are available, which
makes it more difficult to score high on bibliometric evaluations. This implies that bibliometric quality
indicators have limited validity, so that it is less likely that recognition will be based on publications only
and more likely that practical outcomes such as policy advice, patents or spin-off firms count as well.
in urban fields, with a high citation density, bibliometric quality indicators have a higher validity. on
this end of the spectrum these indicators will be more readily used, not only in formal evaluations and
management decisions, but also in informal processes of exchanging recognition.
Explaining variation in the relationship between practical applications and scientific productivityThe variation in the relationship between practical applications and scientific productivity, which
seems crucial for the fate of scientific fields, can be explained by taking into account characteristics of
other actors in the research system, in particular the end-users of academic research. depending on its
cognitive content, each field has different potential users outside university. of particular importance
are ‘upstream end-users’, stakeholders with formal channels to influence the strategies and programs of a
scientific field through research funding, regulation, or policy (lyall et al. 2004 p. 79). examples of such
users are animal breeding firms, environmental policy makers and patient organizations.
in my case studies, the fields that were most successful in combining stakeholder interactions with
academic performance, were fields with wealthy and powerful upstream end-users with a long-term
vision on the utility scientific research (see table 7.5). chemical industry (in the case of catalysis)
and animal breeding industry (aBg) both invest substantial sums in academic research, based on the
expectation that these will pay back on the longer run. These companies support academic researchers
in fundamental research activities, which provide good opportunities for high-impact publications. in
this way, they support scientists along the complete cycle of credibility. The same goes for environmental
policy makers and oil companies in the case of paleo-ecology.
187
conclusions and discussion
Table 7.5 The upstream end-users of different scientific fields. The fields are ranked according to their degree of synergy between practical applications and scientific productivity
Field Upstream end-users
catalysis industry
aBg animal breeding firms
Paleo-ecology Policy makersoil companies
environmental chemistry Policy makersindustryngos
aPS farmers Policy makers
toxicology Policy makersindustryngos
cell biology - (some agro-food companies)
Biochemistry -
Biochemistry and cell biology, in contrast, hardly have any upstream end-users. of course, stakehold-
ers can be identified that may eventually benefit from these research activities, such as patient groups,
farmers or veterinary surgeons. These, however, rather function as ‘downstream’ users, as they are no
active players in the academic research system; they do not directly commission research or influence its
directions. The only actors generously supporting these fields and directly influencing their directions
are research councils (on both national and european level), but these rather function as intermediaries
providing channels to transfer knowledge to and from downstream end-users.
in the third class of fields, including environmental chemistry, toxicology, and aPS, upstream end-
users can be identified as part of the research system, but these mainly support application-oriented
research. These stakeholders definitely care about the research in these areas, but they cannot afford
investments with a long time-horizon. for instance, support from a governmental body for academic
research in the area of toxicology or environmental chemistry is usually connected to a knowledge need
on a specific problem. This explains why researchers in these fields often experience a tension between
end-user relevance and scientific productivity. The short time-horizon of the projects commissioned
by upstream end-users is incompatible with the fundamental nature of dominant debates in scientific
literature. in such cases, interactions with stakeholders catalyze some conversions of credibility (funding
acquisition), but it inhibits others (publishing).
188
chapter 7
another significant variable is the homogeneity of the upstream end-users of a particular field. in the
cases of catalysis and aBg, the set of upstream end-users is quite homogeneous, but in environmental
chemistry, toxicology, and aPS it is heterogeneous. it seems that a homogeneous set of end-users makes
it easier to build a consistent project portfolio, which will help to find synergy between practical applica-
tions and scientific productivity.
Sub-question 3This section shows that a possible explanation for the differences among the changes in the struggles for
relevance observed in different scientific fields can be found in variations in the degree of convergence
in their search pattern, their strategic task uncertainty, their traditional communication culture, and char-
acteristics of their upstream end-users.
General research questioncombining the answers to the three sub-questions, the following answer to the general research ques-
tion of this thesis can be formulated:
In the period 1975-2005, three general trends were visible in the struggles for relevance of Dutch academic
researchers in chemistry, biology and agricultural science: the intensification of the struggle for relevance during
the collection of data, the decreasing value of practical applications as a source of recognition, and the inten-
sification of the struggle for relevance in the context of funding acquisition. However, the changes also varied
across scientific fields, regarding the intensity of stakeholder interactions during data collection, the precise value
of practical applications as a source of recognition, the actual influence of stakeholders on the research agenda,
and the relationship between practical applications and scientific productivity. The changes observed can be
understood as effects of structural changes, in particular the diversification of funding, the rise of performance
evaluations, and changing views on the societal position of the university, in combination with characteristics of
scientific fields and their stakeholders.
7.5 Reflection on central concepts
having answered my research question, let me now reflect on the strengths and weaknesses of the three
central concepts i have used and provide recommendations for further research. in the next sections i
will discuss my contribution to the debate about changing science systems (7.6) and formulate recom-
mendations for science and innovation policy (7.7).
189
conclusions and discussion
The credibility cyclein this thesis the credibility cycle has turned out to be a helpful concept for investigating daily research
practices. My analysis has focused on the reward structure of academic science, rather than on actual
research activities. The credibility cycle proved to be a useful concept for this endeavor, as it guided my
attention to the pressures and incentives that scientists experience in their daily work. in this way, it
helped to get an overview of the general sequence of steps needed in building and conserving a scientific
reputation.
Beside its function as a generic model of science as an ongoing process of resource conversions, the
credibility cycle can also be regarded as a historical-specific description of academic research practices.
Based on my empirical findings i would like to suggest a number of adaptations to the cycle in this sec-
ond function. first, two modifications should be considered regarding the structure of the cycle.
My first recommendation is to include other research outcomes beside scientific articles. today in
some fields patents, prototypes, spin-off firms and consultancy reports count as legitimate outcomes as
well (see also (Packer & Webster 1996)). to what extent these alternative outcomes can be used as a
source of recognition is a further empirical question; my research suggests that in most scientific fields
their yield in terms of recognition is relatively small. in some cases, however, entrepreneurial scientists
can ‘by-pass’ peer recognition, and turn such alternative research outcomes directly into money by
licensing patents, consultancy or selling a spin-off firm to a larger company (see Section 4.5). although
academic entrepreneurship is being stimulated by both public policy and university management, i have
seen only few examples of scientists succeeding to earn substantial income for research activities in this
way.
Second, i suggest broadening the concept of recognition beyond informal acknowledgements to
include a more formal component of ‘recognition’ that accounts for citation scores and the outcomes
of performance evaluations. My case studies have shown that today recognition is not anymore based
on individual assessments of the quality of one’s work, but to a large extent the product of numerical
procedures. Because (informal) reputations are strongly influenced by numerical analyses, it seems not
necessary to add another element of ‘formal reputation’. it seems more fruitful to expand the element
‘recognition’, and make it depend not only on (the number and quality of) articles, but also on citation
scores (like the h-index) and other evaluations. in turn, this composite pool of ‘recognition’ can be used
as a resource for acquiring money.
My outcomes have also provided insights into the processes and norms guiding three specific cred-
ibility conversions. These can be used to further specify and qualify the model:
190
chapter 7
1. in many fields the process of collecting data has expanded to include interactions with potential
knowledge users. in some fields this facilitates efficient and effective data collection, but in others it
is a complicating factor141.
2. The value of articles as a source of recognition has (further) increased, at the expense of practical
applications.
3. in the conversion of recognition to money, promises about potential practical applications have
become an influential factor. Being able to express the future societal benefits that one’s work will
generate, is a valuable asset for today’s scientists. related, contacts with potential knowledge users
are a valuable resource to acquire funding. Moreover, due to the increasing programmatic nature of
public funding for university research, good contacts with other researchers can help a great deal in
acquiring money.
Science-society contractThe idea of a contract between science and society is frequently alluded to in science (policy) studies
literature, but to date it has hardly been given a proper definition and operationalization. This thesis has
further developed this concept into a heuristic for analyzing changing science-society relationships of
specific disciplines in a particular country. This heuristic helped to acquire an overview of the changing
societal position of academic research, or – in other words – the changing meaning of relevance. The
discourse on the science-society relationship influences the actors involved in science. Shared visions
on the identity of science and the legitimacy of public investments in science enable and constrain the
activities of individual researchers, science policy makers, and other sponsors of academic research (see
chapter 3).
in future research the mutual relationships among the three components of the contract (identity,
rationales, conditions) could be further explored. My assumption that compatibility among these three
components is a necessary condition for a stable research system could be empirically tested by analyz-
ing historical episodes in which there are tensions among the three components. another interesting
question is which of the components has the primacy. is the public debate about the societal position
of university research always guided by the rationales relating to expected societal benefits? or can one
also find instances in which desirable conditions are taken as a starting point, and rationales are simply
formulated to legitimize these conditions? in this thesis, the main interest was in the consequences of
the changing contract on research practices, and not on interactions and possible tensions among the
elements within the contract.
141 another recent change in this process, which has received relatively little attention in this thesis, is the increasing value of collaborations during the research process, as large-scale collaborative projects are becoming a salient feature in many fields of science (Penders et al. 2009, vermeulen 2009).
191
conclusions and discussion
another direction for further improvement could be to experiment with differentiating the contract
for individual fields within a discipline. This thesis has shown that even within one discipline there can
be huge differences in expectations about societal benefits generated by scientific fields. for example,
funding for biochemistry is often legitimized by loosely referring to possible long term benefits for
biotechnology or medicine, while funding for environmental is typically justified by pointing to short
term improvement of environmental policy. This thesis has shown that the structuring influence of such
specific promises on the agency of individual scientists is as least as powerful as the influence of generic
promises of scientific disciplines.
Research systemBuilding on existing literature, the notion of a research system has helped to analyze the changing
institutional environment of academic research, in the case of agricultural science (chapter 6). The
institutions in the research system shape certain conversions of the credibility cycle, but they are at the
same time (re)produced by research practices. The most profound changes that i have identified in the
agricultural research system concern shifts in the available funding and the rise of systematic perfor-
mance evaluations.
The research system is sometimes conceptualized in terms of three levels, a top level of government
agencies, an intermediary level of funding bodies, advisory bodies and research organizations, and a
research performance level (van der Meulen & rip 1998, Morris & rip 2006). The prominent position
of the government in such a concept of research system suggests that it has a privileged position as the
central steering actor. however, given the fact that the position of the government varies across fields
and across countries, i prefer a horizontal concept of the science system in which research organiza-
tions are surrounded by a variety of institutions, some of which are governmental policy instruments
and others are not. With this horizontal concept one can avoid the misleading suggestion that top-down
steering is the primary coordination mechanism. it leaves more room for bottom-up forces of research-
ers (individually or collectively) giving shape to their institutional environment.
analogous to my suggestion of a field-specific contract, i would also suggest conceptualizing a
field-specific research system. following roseboom and rutten (1998) in this thesis i have applied
the notion of a research system to agricultural science (see chapter 6). But within this discipline i have
observed considerable variations in the incentives and pressures experienced in different scientific fields.
This variation seemed to relate to a great deal to characteristics of the (upstream) end-users of scientific
fields. due to their differences in cognitive content, fields have different stakeholders with knowledge
needs on a shorter or longer term horizon. Building on literature about sectoral or technological innova-
tion systems, a field-specific notion of the research system could be developed, that takes these charac-
teristics into account. a research system containing the end-users relevant for a specific field gives more
direct insight into the institutional environment of the scientists in this field.
192
chapter 7
7.6 Contributions to the literature about changing science systems
The starting point of this thesis was the scholarly debate about changing science systems. Several bod-
ies of literature claim that the position of universities in modern science systems is changing, and that
academic research is increasingly oriented to producing societal benefits. a number of diagnoses of this
development have been presented that explore its causes and its consequences for the cognitive and
organizational characteristics of scientific knowledge production and for the relations with the outside
world (see chapter 2 for a review). having answered the research questions of this thesis (Section 7.2-
7.4), and having briefly reflected on the main theoretical concepts used (Section 7.5), it is now time to
take stock of my contribution to this debate. i will first discuss a couple of general contributions, and
then give a specific reaction to the concept of Mode 2 knowledge production, which is probably the
most influential of all diagnoses of changing science systems.
in general, my two most important findings in relation to the literature about transformations in the
knowledge infrastructure are:
1. Simultaneous to the increasing pressure for practical utility, dutch university researchers experience
an increasing pressure for productivity, as measured in bibliometric terms.
2. Science is much more diverse than most of this literature suggests.
Potential tension between practical applications and scientific productivityThe common thread in most literature on changing science systems is the increasing orientation of
university research to practical utility. indeed, in my case studies i have found that in many fields interac-
tions between scientists and stakeholders have increased, both in the context of funding acquisition and
during actual research activities. however, my analysis has also explored this development in combina-
tion with another dominant trend with far-reaching consequences, the increasing pressure for scientific
publications. The interplay between these two developments hardly receives any attention in existing
diagnoses of changing science systems, but it is of crucial importance because the growing pressure for
scientific publications potentially inhibits the shift towards contextualized, reflexive or transdisciplinary
research. ironically, this development stems from society’s need for accountability. Systematic perfor-
mance evaluations, developed to address this need (van der Meulen 2008), have strongly increased
the pressure to publish, because of their emphasis on bibliometric quality indicators (Weingart 2005).
although several diagnoses have touched upon the growing pressure for accountability, they have largely
neglected this effect. Books on Post-academic science (Ziman 2000), Mode 2 (gibbons et al. 1994) and
academic capitalism (Slaughter & leslie 1997) all report society’s increasing need to have ‘value for
money’, but they only discuss the consequences of this pressure in terms of the increasing orientation to
practical utility and the growing competitiveness of funding acquisition.
193
conclusions and discussion
This thesis has shown that there is a potential tension between the increasing pressure to publish
and the increasing stress on generating broader impacts. in some fields, such as catalysis and aBg, i have
observed a synergetic relationship between societal impacts and scientific excellence. in other fields,
however, like biochemistry and toxicology, the pressure for academic productivity is at odds with the
pressure of practical applications. in these fields scientists have increased their efforts to produce papers
in high-impact journals at the expense of the practical utility of their work. in these areas research activi-
ties addressing knowledge needs of societal stakeholders are not easily published in scientific journals.
here the increased publication pressure inhibits the shift towards application-oriented research modes.
to my knowledge, of all diagnoses of transformations in the science system reviewed in chapter 2, only
the triple helix corpus explicitly addresses this potential tension142.
Diversity of scienceThis thesis shows that the dynamics of the relationships between academic research and its societal en-
vironment vary strongly across scientific fields. in this respect my work adds to a number of other recent
studies that have reported varying reactions to institutional changes across scientific disciplines (albert
2003, gläser et al. forthcoming, reale & Seeber forthcoming). This thesis confirms their call for disci-
plinary differentiation in science (policy) studies. Moreover, it reinforces it to a call for an even more
fine-grained perspective that does not only distinguish among complete disciplines, but also among
specific fields. Some of the diagnoses of changing science systems differentiate across scientific fields
or disciplines, in particular literature about post-normal science, finalization science, triple helix and
innovation systems. however, none contains a satisfying framework to understand the varying dynamics
of scientific fields.
literature about post-normal science distinguishes categories of ‘problem solving’. funtowicz and
ravetz (1993) distinguish ‘core science’ (pure or basic), applied science, ‘professional consultancy’,
‘post-normal science’, with increasing decision stakes and increasing systems uncertainties. My findings
could stimulate this literature to further explore the stakeholder characteristics of different scientific
fields, as the inclusion of users in ‘extended peer review’ will be much easier in fields with clearly identifi-
able upstream end-users.
‘finalization theory’ would explain changes in the struggle for relevance by the internal dynam-
ics of scientific fields. according to finalization theory scientific fields undergo a generic development
from an explorative to a paradigmatic and a post-paradigmatic phase (Böhme et al. 1983). Their claim is
that once a field has matured and sufficient consensus has been attained about the ruling paradigm, the
optimal conditions are fulfilled for fruitful interactions with firms or other knowledge users. This reason-
142 for example, an empirical analysis of university-industry-government relations in South-Korea has shown that govern-mental policy to stimulate publications in international scientific journals has contributed to the erosion of the synergy in triple helix relations (Park & leydesdorff 2010).
194
chapter 7
ing, however, is in contradiction with my finding that relatively young field of aPS has developed more
intensive stakeholder relationships than mature fields like cell biology and biochemistry.
Most of the literature about innovation systems and triple helix interactions is of a generic nature
and does not distinguish among industrial sectors or scientific fields (e.g. Motohashi (2005)), but these
bodies of literature also contain studies that show that in some sectors there are stronger links between
academic science and industry than in others. Their explanations are not restricted to characteristics of
the scientific fields involved, but also involve economic and social characteristics of the industrial (or
governmental) organizations. for example, a triple helix study on micro-electronics showed a syner-
getic relationship between scientific productivity and industrial collaborations similar to my case studies
of catalysis and aBg (Balconi & laboranti 2006). The authors explain this (partly) by the close social
relationships between firms and academic research groups, and by the industrial need for recruitment of
productive researchers. an important limitation of triple helix and innovation system literature is that
both tend to ‘black-box’ research practices. The understanding i have gained of the incentive structure of
university research could help them to improve the understanding of why science-industry relationships
are much more successful in some fields than in others.
in contrast with these four diagnoses, The new Production of Knowledge (nPK), academic
capitalism and Post-academic Science lack a proper differentiation among scientific fields. Their main
concern seems the identification and investigation of general phenomena rather than exploring and
explaining the differences among fields. nPK (gibbons et al. 1994) only contains generic claims; it does
not systematically differentiate across scientific fields. its arguments mainly deal with science, technol-
ogy and medicine; one chapter of the book makes a comparison with the (aggregated) humanities. The
authors use examples from different scientific fields, these all serve the same general argument. Ziman’s
book about Post-academic Science (2000) also has a rather loose empirical foundation and does not
differentiate among scientific fields or disciplines. ‘academic capitalism’ (Slaughter & leslie 1997)
presents data on specific departments representing different fields/disciplines, but the authors hardly
elaborate on their differences. altogether these three books raise the suggestion that all fields of science
are subject to a generic shift, and that there are only differences in degree and not in kind.
in conclusion, the debate about changing science systems can benefit from my findings regarding the
different dynamics of scientific fields. This thesis has shown how, under similar institutional conditions,
some fields develop a stronger practical orientation, while others strengthen their academic identity.
By providing an explanatory framework for this variation based on characteristics of fields and their
stakeholders (Section 7.4), this thesis contributes to the development of a more differentiated view on
the changing relationship between science and society.
195
conclusions and discussion
A shift towards Mode 2 knowledge production?The diagnosis about ‘Mode 2 knowledge production’ (gibbons et al. 1994) deserves special attention,
because it is probably the most influential writing on changing science systems of the past few de-
cades143. our systematic literature study raised the suggestion that Mode 1 and Mode 2 should be seen
as ideal-types rather than categories of existing research practices (chapter 2). What is more, because
of the limited coherence of the five major attributes of Mode 2 (see table 7.6), it seemed appropriate
to investigate them separately. although the central research questions of this thesis were not expressed
in terms of Mode 2 attributes, my findings give an indication regarding the degree to which they have
changed in the netherlands. in short, my empirical results confirm two of the trends announced by gib-
bons et al. (1994), but challenge the three others, in particular the claim that quality control is increas-
ingly dominated by societal criteria.
Table 7.6 results in terms of Mode 2 attributes
Attribute Status in literature*
Trend in my case studies
context of application Supported increasing in most fields
heterogeneity Supported increasing in most fields
transdisciplinarity contested increasing, but still marginal, due to lack of incentives and rewards
reflexivity contested increasing, but accompanied by increasing scientific account-ability
novel quality control contested increasing in funding distribution, but not in other processes*see chapter 2
in agreement with other literature, my case studies indicate that in most fields the ‘heterogeneity’ of
knowledge production is increasing, and that a growing share of it is conducted in the ‘context of ap-
plication’. in most fields knowledge production is increasingly organized around possible practical
applications. for example, research in paleo-ecology is more and more inspired by knowledge needs
of environmental policy, and animal breeding & genetics research is increasingly oriented to industrial
applications. This trend seems mainly driven by shifts in available funding. it also relates to the increas-
ing heterogeneity of knowledge production i observed in most fields. Beside academic researchers,
industry, policy makers and other (possible) knowledge users make active contributions to the research
process. in funding arrangements, such as the dutch technology foundation StW and technological
top institutes, users committees are set up to give feedback on the progress of individual projects or
143 currently ( June 2010) ‘The new Production of Knowledge’ (gibbons et al. 1994) has been cited almost 1900 times, and its popularity seems still increasing (hessels & van lente 2010a)
196
chapter 7
aggregated programs. My interview data suggest that such input significantly contributes to the produc-
tion of knowledge. also, in aBg and catalysis, a significant share of all research activities is organized in
teams including both academic and industrial researchers. in other fields, like aPS and environmental
chemistry, active contributions of researchers from public research institutes are increasingly essential to
the research process.
however, the rise of the other three attributes of Mode 2, transdisciplinarity, reflexivity, and novel
quality control, has less support in the literature (see chapter 2). My case studies provide further
reasons to doubt the idea of a homogeneous rising trend of these characteristics. regarding all three at-
tributes, my empirical data indicate a mixed picture: a certain increase is visible, but it is accompanied by
an undercurrent in the opposite direction, most significantly regarding quality control.
My findings do not support the view of a general increase of the dynamic integration of elements of
several disciplinary backgrounds. Transdisciplinarity can only clearly be recognized in one of my cases,
animal production systems (aPS). here i observed that novel combinations are made of methods and
theories from other disciplines, in order to develop knowledge to support sustainable animal produc-
tion. researchers develop ‘distinct theoretical structures, research methods and modes of practice’
(gibbons et al. 1994 p. 168) that can not be easily located on the prevailing disciplinary map. in a
weaker sense, some of these aspects have also become visible in aBg, toxicology and paleo-ecology, but
for this type of research ‘multidisciplinarity’ or ‘interdisciplinarity’ seem more appropriate terms than
‘transdisciplinarity’. in these fields the integration of elements from different disciplinary backgrounds is
less dynamic and interactive than in aPS. toxicologists collaborate increasingly with medical scientists;
paleo-ecologists with geo-scientists, but their resulting approaches seem still reducible to their disci-
plinary parts. in correspondence with suggestions in the literature (Merkx & van den Besselaar 2008,
Kaufmann & Kasztler 2009), the academic reward system does not stimulate transdisciplinarity in aPS.
The pressure for academic publications stemming from dominant mechanisms of quality control tends
to work as an inhibitor for transdisciplinary research, as this does not typically result in prestigious jour-
nal papers. altogether, my case studies suggest that the share of transdisciplinary research is growing, but
that it is far from a mainstream approach in dutch university research.
With regard to reflexivity (or ‘social accountability’), my case studies indicate an ambivalent develop-
ment. Judging from my interviews, in general the reflexivity of dutch academic researchers seems to
have increased over the past 30 years. due to the need for promising ‘relevant’ contributions in order to
get funding and the shifting societal views on academic research, scientists have become more sensitive
to the broader implications of their work. regardless of the basic or applied nature of his/her research,
a certain awareness of potential applications has become a valuable asset for every academic scientist.
even most biochemists, who interact little with societal stakeholders and whose work does not regularly
lead to direct policy solutions or technological innovations, nowadays have a view on the societal impact
197
conclusions and discussion
their work could have on the long run144. The reflexivity is obviously much higher in fields that are
more strongly connected with practical applications, such as aPS, paleo-ecology and catalysis. in these
fields, a consciousness of possible impact is continuously present, and this is stimulated by the frequent
interactions with knowledge users in industry or the government. in more fundamentally oriented fields,
like biochemistry and cell biology, such consciousness is present mainly during the funding acquisition
phase, and fades away once the actual research work is started.
The increased reflexivity i have observed can be understood as a response to a societal pressure
for accountability. My three case studies indicate that society has developed a need for more reflexive
science, science that is aware of (or even anticipates on) its own intended and unintended (side-)effects.
The increasing policy steering of research priorities and stimulation of interactions with stakeholders can
(also) be seen as a reaction to public concerns about undesirable side-effects of technoscientific develop-
ment. The effect of these policies can be interpreted as a de-differentiation of different societal realms
(nowotny et al. 2001). Beside scientists, academic researchers increasingly take up the roles of public
experts and of entrepreneurs145.
Yet, my research shows that the (growing) reflexivity of scientists is limited by characteristics of
the research system, especially the pressure for high scientific productivity. This pressure has led to an
increasing scientific accountability, that is, awareness of the scientific value of academic research. The
increased pressure for excellent academic performance has incited an increased consciousness of pos-
sibilities to turn one’s outcomes into scientific publications, preferably in high-impact journals. This
parallel development should definitely be taken into account when discussing the increasing reflexivity,
especially because in some cases (like toxicology and cell biology) the scientific accountability seems to
have increased much more than the social accountability.
The strongest disagreement between my empirical results and the Mode 2 diagnosis concerns the
question of changing academic quality control. My findings contradict the claims by gibbons et al.
(1994) and by hemlin and rasmussen (2006) that academic quality control is increasingly ruled by
societal considerations and conducted by non-scientific actors. in most fields, the various mechanisms
of academic quality control turn out to have developed in different, partly opposite directions. Similar to
144 e.g. ‘With this method you can look at protein quantities in response to external changes. eventually you can sum-marize this in a computer model. So then you actually have a sort of model of a cell. and if you administer drugs, then what will happen in the cell? Maybe you can then predict what effects it has. With a model you know what processes correlate within a cell. So when something changes, then this will change and that. if you develop a certain drug against a particular disease and you administer it, then you assume that it only has the effect you are aiming for. But it could also influence other processes. and that is what you could predict with such a model.’ (researcher c4)
145 although these new roles certainly enhance the reflexivity of scientists, because they widen their perspective, they can also create new side-effects, due the new stakes and interests that scientists develop. active engagement in commercial activities or policy advice can threaten the disinterestedness of scientists (Ziman 2003, Jacob forthcoming). Because of their personal stake in particular research outcomes, the objectivity of their research could be in danger and this could eventually lead to undesirable outcomes, like technologies carrying unacceptable risks to human health or the environ-ment.
198
chapter 7
Schmoch and Schubert (forthcoming), i found that the different practices of quality control each have
their own criteria, which are often contradictory. in agreement with the claims by gibbons et al., the
procedures used in the distribution of research funding are multidimensional and increasingly use valo-
rization, societal relevance or equivalents as a selection criterion. however, job interviews, performance
interviews and performance evaluations most strongly reward scientific productivity and scientific
impact. in these practices of quality control, a narrow set of (disciplinary) scientific criteria dominate,
in particular publications numbers, journal impact-factors, and citation scores. altogether my findings
do not indicate that in the netherlands societal considerations and users are increasingly powerful in
academic quality control. Peer review remains a dominant steering mechanism controlling problem
choice, career advancement and reputations. The wider set of criteria and broader social composition of
the review system that gibbons et al. predicted have only become visible in the mechanisms control-
ling who receive funding, but not in mechanisms controlling what precise topics are investigated, what
research methods are used or what papers are published. in short, nowadays societal considerations do
play a significant role in quality control in the funding phase, but not during or after the actual research
phase.
in conclusion, my empirical findings regarding Mode 2 further substantiate the theoretical criticisms
uttered in chapter 2. The fieldwork reported in this thesis shows that the dynamics of contemporary sci-
ence systems are neither homogenous nor unidirectional. The issues of reflexivity and quality control are
ambivalent and vary strongly across scientific fields. Based on these findings i conclude that the idea of a
homogeneous transformation of the science system is misleading.
7.7 Policy recommendations
The research reported in this thesis was primarily a scholarly endeavor. My first aim was not to assess the
effectiveness of science and innovation policies, but rather to enhance the understanding of changing
science systems. Still, given the topic of this thesis my analysis dealt in quite some detail with policy
instruments like funding mechanisms and evaluation processes, and with their effects on academic
research practices. for this reason, it seems appropriate to reserve some space for a consideration of the
practical implications of my findings.
The two most salient developments in the governance of academic research discussed in this thesis
are the pressure for valorization related to the diversification of funding and the rise of performance
evaluations. in the following i will make a number of critical remarks on both topics. With regard to
funding, i comment on the dangers of a restrictive notion of ‘valorization’. concerning evaluations,
i point to the dangers and limitations of bibliometric quality indicators. i will conclude with a list of
recommendations.
199
conclusions and discussion
however, i start with a more general statement. for a long time, the distinction between basic and
applied research has served as a landmark in science and innovation policy. nowadays this dichotomy is
denied in most policy documents. This denunciation is inspired by literature about Mode 2 knowledge
production, strategic science and Pasteur’s Quadrant. These contributions claim that fundamental and
applied research are not mutually exclusive and that research projects can be perfectly fundamental
and applied at the same time. however, my results indicate that in many fields there is still a tension
between enhancing fundamental understanding and generating practical benefits, and in some fields this
tension is pretty strong (e.g. in environmental chemistry). The idea that these two go very well together
in all fields of science may be wishful thinking, based on the extrapolation of observations on a limited
number of fields such as catalysis or biotechnology. May this thesis stimulate awareness of the possible
tension between practical applications and scientific excellence: a strong pressure for valorization can
threaten the viability of fundamental research lines and a strong pressure for academic excellence can
threaten the alignment of research activities with stakeholder needs. This general concern forms the
starting point for the more specific considerations below.
Knowledge valorizationas my analysis has shown, the latest conception of the relevance of university research in the nether-
lands is strongly connected with the idea of ‘knowledge valorization’ (see table 7.1). This term refers to
‘making sure that research results are being applied and converted into economic, financial and societal
value’146. nowadays valorization figures as a goal in a large share of public funding arrangements for
dutch university research147.
in principle, knowledge valorization seems a perfectly legitimate goal. as academic research is still
paid predominantly from public money, the generation of societal benefits seems a logical requirement.
valorization as defined above is so general that one can hardly object to it. it is even similar to ‘relevance’
in the sense that actors can give it different meanings, according to their own preferences and interests.
however, there is a tendency to restrict the meaning of valorization by placing specific emphasis on
economic value, generated on the short term (Benneworth & Jongbloed 2010). for example, the Ministry
of education, culture and Science defines valorization as ‘the conversion of research results into eco-
nomic value’148. The restriction to short term economic contributions carries four dangers.
first this limited notion of valorization suggests that the specific benefits of university research can
be predicted, or even planned. This perspective is incompatible with the view of most researchers i
146 vSnu, 2005, ‘onderzoek van waarde: activiteiten van universiteiten gericht op kennisvalorisatie’, utrecht.147 in its most recent strategy paper, research council nWo has announced that it will pay attention to possible ‘knowl-
edge utilization’ (which is essentially a synonym for valorization) within all research it is funding (nWo, 2010, ‘groeien met kennis: Strategienota nWo 2011-2014’, den haag)
148 Ministry of education, culture and Science, 2004, ‘focus op excellentie en meer waarde: Wetenschapsbudget 2004’, den haag., p. 12
200
chapter 7
have interviewed, who experience writing valorization plans (and the like) as window-dressing149. Most
procedures for acquiring (application-oriented) funding from public sources do not do justice to the
diversity and unpredictability of science. Many of the programs of the dutch research council nWo and
european framework Programmes require detailed projections of the expected societal benefits or even
a plan or strategy to ensure such benefits are realized. in some fields it seems appropriate to anticipate on
collaborations with specific firms or the development of a particular product, but in others this does not
make sense.
Second, the value of academic patents is systematically overrated. university researchers experience
a pressure from their employers and by research councils to patent potentially commercially relevant
findings. however, my impression is that this is driven by the need of these organizations to legitimate
their own societal position, rather than by truly financial motives. academic patents are often regarded
as an indicator for successful valorization. as a result, many patents may be filed for symbolic reasons150.
very few of my respondents received significant income from patents. This fits the global picture in
which only a few universities manage to attract substantial revenues from patents; for most universities
patenting is not profitable (geuna & nesta 2006).
Third, the emphasis on short term benefits endangers the viability of research lines oriented to
fundamental understanding. valorization policy tends to restrict the autonomy of scientist to choose
their own research questions and approaches. My fieldwork shows that in some fields the demand for
practical applications leads scientists away from fundamental questions. in my opinion there are at least
three reasons to protect fundamental research lines:
1. in spite of the current hype of high-tech spin-off companies, the most effective channel for knowl-
edge valorization probably remains the training of new generations of scientists (Salter & Martin
2001). Scientists trained in basic research are a crucial requirement for innovative performance. if
all Phd-projects are tied to practical goals, the new generation of researchers could lack a basis of
fundamental understanding, especially in fields where there is a tension between practical applica-
tions and scientific excellence.
2. one of the crucial functions of academic research activities is securing the quality of higher educa-
tion programs by keeping ‘teachers’ at the forefront of academic debates. it strikes me that the
education function of universities receives little attention in the current valorization discourse. Some
research areas which are not particularly promising in terms of generating direct social or economic
impacts are highly relevant with regard to education, most strikingly in the social sciences and the
humanities. on the longer run, too much steering on valorization could endanger the quality of
higher education in these areas.
149 This was also shown by Morris (2000) and roberts (2009)150 a survey among scientists at german Max Planck institutes has shown that their patenting behavior is more strongly
influenced by a quest for reputation than by expectations of real financial benefits (göktepe-hulten & Mahagaonkar forthcoming)
201
conclusions and discussion
3. Third, basic research also has a cultural function. in the numerous conversations i have had about
the societal relevance of science with people inside and outside the science system i noticed that
there is widespread sense that enhancing the understanding of matter, life, or the universe is part of
our culture, regardless of its practical benefits151.
The fourth danger of the restricted concept of valorization relates to the importance of external
funding sources for the survival of research groups. The overall decrease in block grant support for
university research threatens the stability of academic research groups and the coherence of their activi-
ties. research groups now have to combine a diverse collection of funding sources, each with their own
demands. given the limited total sum of public funding available for academic research, many scientists
depend to a large extent on external, often commercial funding sources. however, empirical evidence
indicates that industrial funding is not a decisive factor for successful valorization152. in some fields (like
toxicology and environmental chemistry) i have found instances where external sponsors threaten the
independent position of university researchers. commercial sponsors often delay the free communica-
tion of research results. Moreover the integrity of scientists is endangered in the case their sponsors have
a strong stake in particular research outcomes.
in conclusion, a restricted concept of valorization is a danger for the long term viability of science, as
it threatens several of its vital functions. on the longer run, this perspective impedes rather than stimu-
lates the conversion of research results into economic and societal value. truly successful valorization
requires a long-term horizon. Because most societal benefits become visible long after research projects
are finished, it makes more sense to govern by strategic priorities than by immediate impact.
Bibliometric quality indicatorsThe second issue that i want to comment on is the dominance of bibliometric quality indicators in scien-
tific performance evaluations. in my fieldwork i observed that bibliometric data (like publication num-
bers, impact factors, citations) enjoy an ever growing popularity as indicators of the quality of individual
scientists, organizations or journals. When used with care, bibliometric analysis can be a valuable tool to
evaluate the performance of individuals, research groups or institutes. Bibliometric indicators have been
adopted so abundantly both in formal and informal quality assessments, without proper acknowledge-
ment of their limitations, that they have radically changed the meaning of a scientific paper. originally
a means of communication, publications have become an end in itself. This development carries at least
three dangers:
151 a possible fourth reason for cherishing fundamental research is that autonomous fundamental research can also lead to unpredictable outcomes with a social or economic impact that is much higher than research that was originally de-signed to contribute to particular practical goals. however, because this effect is difficult to prove, it is difficult to draw conclusions about the amount of basic research needed to this end.
152 Between 1997 and 2004 Stanford university, generally considered as one of the most successful examples of academic entrepreneurship, has received less than 5% of its funding from industry (tindemans & Soete 2007).
202
chapter 7
first, i have seen that bibliometric indicators tend to restrict the concept of academic excellence.
The excessively high value attributed to numbers of journal papers and citations has left little room for
other aspects of academic performance, in particular teaching and knowledge transfer or dissemination
activities (hessels & van lente 2010b). also, applied research and interdisciplinary research (even if it
is of a rather fundamental nature) is undervalued, because it usually yields fewer papers in high-impact
journals.
The second risk of bibliometric quality indicators is the promotion of a monoculture of science.
across all fields, the scientists i have interviewed experienced a pressure to assimilate to the ‘urban’
communication style (Becher & trowler 2001) that in principle fits only a limited share of the sciences.
The generic incentives for more publications in scientific journals endanger fruitful publication tradi-
tions in some fields that involve more book publications and papers in native languages (like animal
production systems). forcing these fields to assimilate to the dominant publication form could increase
the efficiency of their communication mechanisms, but it threatens the relevance of their work for local
stakeholders and it could also threaten the viability of larger (and more risky) research projects, the
results of which can not be easily be communicated in brief journal papers.
Third, the high pressure for publications creates an inflation effect on the value of a scientific paper.
in the pressure to publish the sheer number of papers often seems more important than their precise
content or quality. Several respondents have reported to divide their results into the ‘smallest publish-
able units’ to enhance their publication rate153. Sometimes the same findings can be published in several
different journals, if the author manages to opportunistically emphasize different aspects of the same
study for different audiences. This trend probably decreases the efficiency of the scientific communica-
tion system.
to conclude, bibliometric quality indicators are useful tools in principle, but their universal applica-
tion has several perverse side-effects. This problem can not be solved by only developing more sophisti-
cated indicators, it is also necessary to reduce their general influence on decisions in the academic world.
i see two possibilities to do so: complementing quantitative bibliometric indicators with qualitative as-
sessments, and complementing them with indicators of social and economic impact.
153 My qualitative observations on this point can be supported with quantitative evidence from australia, where the introduction of performance-based funding formulae resulted in an increase of the average productivity, but a decrease in the average impact of the papers published (Butler 2003).
203
conclusions and discussion
Practical recommendationsi finish with two sets of recommendations that may help to alleviate the problems identified.
Regarding research funding:
1. loosen the ‘valorization’ requirements for grants from the national research council nWo. it makes
sense to stimulate applicants to think about possible practical applications, but not to force them to
plan specific knowledge dissemination activities, in a phase they don’t know yet what type of know-
ledge they will generate.
2. Make a more careful cost-benefit analysis before patenting academic inventions. do not use patents
as an exclusive indicator of valorization success. in most cases, commercially valuable inventions can
better be sold directly to an interested company rather than patented.
3. The governance of academic research activities should be based on expected long-term benefits
rather than short-term benefits. organize processes of strategic foresight in order to explore and
choose long-term priorities for scientific research154. Such exercises should be protected from do-
minant incumbent forces, in order to prevent the reproduction of existing (economic) powers into
future priorities. Moreover consistency should be strived for, by setting strategic goals for periods
beyond the regular term of a governmental administration.
4. restore and protect a healthy share of unconditional block grant support for university research. a
further decrease of this money flow would endanger the long-term viability of the dutch science
system.
Regarding research evaluation:
1. use bibliometric indicators with care. in any quality judgment (whether dealing with individuals,
organizations or journals), bibliometric indicators should be accompanied by a qualitative analysis.
a direct linkage between funding and bibliometrics should be avoided under all circumstances.
Bibliometric quality indicators have important methodological limitations, especially when evaluat-
ing individuals and when comparing across fields. What is more, if funding decisions are left to
automated procedures a crucial opportunity for reflection is missed. decisions about the future
should not be based only on past successes but also on a consideration of future priorities, based on
foresight studies and normative goals.
2. Broaden the criteria of performance evaluations. as a recent series of pilot studies has shown, there
are also possibilities to measure social and economic impact of academic research155. as most soci-
etal benefits only become visible on the longer run, such indicators are not applicable to all scientific
154 inspiration can be drawn from the uK foresight Programme, which conducts in-depth studies 20-80 years into the future (http://www.foresight.gov.uk).
155 eric 2010, Handreiking Evaluatie van maatschappelijke relevantie van wetenschappelijk onderzoek, eric publicatie 1001
204
chapter 7
fields. however, in numerous fields they can have an equal position next to scientific quality, produc-
tivity and relevance.
3. Broaden the criteria for the promotion and hiring of academic staff. They should be less dependent
on quantitative indicators of a candidate’s research performance, and more on a qualitative assess-
ment of the content of his/her work, including any societal impacts. in addition teaching qualities
deserve an equal position next to research qualities in such decisions. This could stimulate scientists
to invest more time and energy in teaching.
7.8 Conclusion
When and how can science be called relevant for society? Science promises considerable societal
benefits: it will help to improve public health, enhance global security and protect the environment. Still
the outcomes of research activities are inherently uncertain and unpredictable. how to govern science
in order to optimize its relevance for society? The understanding of the complex relationships between
academic research and its practical applications is limited. What is more, the question of relevance is
particularly urgent against the background of the current debate about changing science systems.
in this thesis i have studied the various struggles for relevance of university researchers, in other
words, their efforts to make their work correspond with ruling standards of relevance, and to influence
these standards. i have conceptualized these struggles as part of the credibility cycle. The agency of sci-
entists in this cycle is influenced by structural conditions, which can be seen in terms of a science-society
contract or as a research system. My empirical investigations have shown how struggles for relevance
have changed over the past 30-35 years in eight fields of natural science in the netherlands and provided
starting points for explaining the changes observed.
across all scientific fields, i observed three common developments: the intensification of the
struggle for relevance during the collection of data, the decreasing value of practical applications as
a source of recognition, and the intensification of the struggle for relevance in the context of funding
acquisition. These general trends can understood as effects of the combination of three major changes in
the structural conditions of academic research in the netherlands: the diversification of funding, the rise
of performance evaluations, and changing views on the societal position of the university.
The changes in the struggle for relevance also varied across scientific fields, along four dimensions:
the intensity of stakeholder interactions during data collection, the precise value of practical applica-
tions as a source of recognition, the actual influence of stakeholders on the research agenda, and the
relationship between practical applications and scientific productivity. a possible explanation for these
differences can be found in the combination of the degree of convergence in their search pattern, their
205
conclusions and discussion
strategic task uncertainty, their traditional communication culture, and characteristics of their upstream
end-users.
This study has two major implications for the debate about changing science systems. first, it shows
that the increasing pressure for productivity, as measured in bibliometric terms, can counteract the pres-
sure for practical utility. in many fields i observed a tension between research activities that are useful for
societal stakeholders and activities that are most promising in terms of scientific publications. Second,
my work indicates that a further differentiation is needed, as the dynamics of science vary much more
across scientific fields than most literature suggests. Both findings challenge the empirical validity of
some central claims around the notion of Mode 2 knowledge production, in particular regarding trans-
disciplinarity, reflexivity, and novel quality control.
Both the meaning of relevance and the ways in which scientists struggle for relevance are contingent;
they change over time and vary across scientific fields. This thesis has mapped some of these changes and
variations, and the tensions they may create. By providing conceptual clarity and empirical insights, this
thesis contributes to the scholarly debate about changing science systems. on the longer run, i hope this
thesis will, directly or indirectly, also contribute to the possibilities of science governance to effectively
increase the societal benefits of science. after all, academic research is intimately entangled with rel-
evance, and it probably always will be.
207
references
albert, M. 2003, universities and the market economy: The differential impact on knowledge produc-
tion in sociology and economics, Higher Education, 45 (2): 147-182.
aWt 2004, De prijs van succes - over matching van onderzoekssubsidies in kennisinstellingen, aWt advies
58, april 2004, den haag.
Bakker, S., van lente, h. & Meeus, M. 2008, arenas of expectations for hydrogen technologies, ISU
Working Paper, #08.19.
Balconi, M. & laboranti, a. 2006, university–industry interactions in applied research: The case of
microelectronics, Research Policy, 35 (10): 1616-1630.
Baldursson, e. 1995, The Elusive Frontier: On the emergence and change of a science-society contract, Phd
thesis, göteborg university, göteborg.
Barker, K. 2007, The uK research assessment exercise: the evolution of a national research evaluation
system, Research Evaluation, 16 (1): 3-12.
Becher, t. & trowler, P.r. 2001, Academic Tribes and Territories, Srhe and open university Press,
Maidenhead, Berkshire.
Benneworth, P. & Jongbloed, B. 2010, Who matters to universities? a stakeholder perspective on huma-
nities, arts and social sciences valorisation, Higher Education, 59: 567-588.
Benner, M. & Sandstrom, u. 2000, institutionalizing the triple helix: research funding and norms in the
academic system, Research Policy, 29 (2): 291-301.
Bieleman, J. (ed.) 2000, Landbouw, Part I (p.11-233) in ‘Techniek in Nederland in de Twintigste Eeuw:
Landbouw en Voeding’, Walburg Pers, Zutphen.
Blume, S.S. 1985, The development of Dutch science policy in international perspective, 1965-1985, raad van
advies voor het Wetenschapsbeleid, Zoetermeer.
Blume, S.S. & Spaapen, J.B. 1988, external assessment and “conditional financing” of research in dutch
universities, Minerva, 26 (1): 1-30.
208
references
Böhme, g., van den daele, W., hohlfeld, r., Krohn, W. & Schäfer, W. 1983, Finalization in Science: The
Social Orientation of Scientific Progress, riedel, dordrecht.
Böhme, g., van den daele, W. & Krohn, W. 1973, die finalisierung der Wissenschaft, Zeitschrift für
Soziologie, 2 (2): 128-144.
Bonaccorsi, a. 2008, Search regimes and the industrial dynamics of science, Minerva, 46: 285-315.
Boomkens, r. 2008, Topkitsch en Slow Science, van gennep, amsterdam.
Boon, W. & Broekgaarden, r. 2010, The role of patient advocacy organisations in neuromuscular disease
r&d - The case of the dutch neuromuscular disease organisation vSn, Neuromolecular Disorders,
20: 148-151.
Borrás, S. 2006, legitimate governance of risk at the eu level? The case of genetically modified orga-
nisms, Technological Forecasting and Social Change, 73 (1): 61-75.
Bos, B. & grin, J. 2008, “doing” reflexive modernization in pig husbandry: The hard work of changing
the course of a river, Science, Technology and Human Values, 33 (4): 480-507.
Bourke, P. & Butler, l. 1998, institutions and the map of science: Matching university departments and
fields of research, Research Policy, 26 (6): 711-718.
Bozeman, B. & corley, e. 2004, Scientists’ collaboration strategies: implications for scientific and techni-
cal human capital, Research Policy, 33: 599-616.
Braun, d. 2003, lasting tensions in research policy-making - a delegation problem, Science and Public
Policy, 30 (5): 309-321.
Braun, t. & Schubert, a. 2003, a quantitative view on the coming of age of interdisciplinarity in the
sciences 1980-1999, Scientometrics, 58 (1): 183-189.
Bruce, a., lyall, c., tait, J. & Williams, r. 2004, interdisciplinary integration in europe: the case of the
fifth framework programme, Futures, 36: 457-470.
Bush, v. 1945, Science: The endless frontier.
Butler, l. 2003, explaining australia’s increased share of iSi publications: the effects of a funding for-
mula based on publication counts, Research Policy, 32 (1): 143-155.
carlsson, B. & Stankiewicz, r. 1991, on the nature, function and composition of technological systems,
Evolutionary Economics, 1: 92-118.
cassidy, P.J. & radda, g.K. 2005, Molecular imaging perspectives, Journal of the Royal Society Interface, 2
(3): 133-144.
chesbrough, h.W. 2003, Open Innovation: The New Imperative for Creating and Profiting from Technology,
harvard Business School Press.
cooke, P., gomez uranga, M. & etxebarria, g. 1997, regional innovation systems: institutional and
organisational dimensions, Research Policy, 26 (4-5): 475-491.
209
references
cooper, M.h. 2009, commercialization of the university and problem choice by academic biological
scientists, Science Technology and Human Values, 34 (5): 629-653.
costas, r. & Bordons, M. 2007, The h-index: advantages, limitations and its relation with other biblio-
metric indicators at the micro level, Journal of Informetrics, 1 (3): 193-203.
craye, M. & funtowicz, S. 2005, a reflexive approach to dealing with uncertainties in environmental
health risk science and policy, International Journal of Risk Assessment and Management, 5 (2):
216-236.
de Boer, h., enders, J. & leišytė, l. 2007, Public sector reform in dutch higher education: the organiza-
tional transformation of the university, Public Administration, 85 (1): 27-46.
de Wit, J., dankbaar, B. & vissers, g. 2007, open innovation: the new way of knowledge transfer? Jour-
nal of Business Chemistry, 4 (1): 11-19.
despres, c., Brais, n. & avellan, S. 2004, collaborative planning for retrofitting suburbs: transdisciplina-
rity and intersubjectivity in action, Futures, 36: 471-486.
dijksterhuis, f.J. & van der Meulen, B. 2007, Tussen Coördineren en Innoveren: De Nationale Raad voor
Landbouwkundig Onderzoek, 1957-2000, nederlands agronomisch historisch instituut, gronin-
gen.
dijstelbloem, h., Spaapen, J. & Wamelink, f. 2002, Knocking at the gatekeepers door: the distribution
of accountability in pharmaceutical research, Society for Social Studies of Science (4S) conference,
Milwaukee, uSa.
duffus, J.h. 1980, Environmental Toxicology, edward arnold, london.
edquist, c. 1997, Systems of Innovation: Technologies, Institutions and Organisations, Pinter publishers,
new York/london.
eisenhardt, K.M. & graebner, M.e. 2007, Theory building from cases: opportunities and challenges,
Academy of Management Journal, 50 (1): 25-32.
elzinga, a. 1997, The science-society contract in historical transformation: with special reference to
“epistemic drift”, Social Science Information, 36 (3): 411-445.
etzkowitz, h. 2003, research groups as ‘quasi-firms’: the invention of the entrepreneurial university,
Research Policy, 32: 109-121.
etzkowitz, h. 1998, The norms of entrepreneurial science: cognitive effects of the new university-indus-
try linkages, Research Policy, 27: 823-833.
etzkowitz, h. & leydesdorff, l. 2000, The dynamics of innovation: from national Systems and ‘’Mode
2’’ to a triple helix of university-industry-government relations, Research Policy, 29 (2): 109-123.
etzkowitz, h. & leydesdorff, l. 1998, The endless transition: a “triple helix” of university-industry-
government relations, Minerva, 36: 203-208.
210
references
etzkowitz, h., Schuler Jnr, e. & gulbrandsen, M. 2000a, “The evolution of the entrepreneurial univer-
sity” in The Future of Knowledge Production in the Academy, eds. M. Jacob & t. hellström, Srhe
and open university Press, Buckingham, p. 40-60.
etzkowitz, h., Webster, a., gebhardt, c. & terra, B.r.c. 2000b, The future of the university and the
university of the future: evolution of ivory tower to entrepreneurial paradigm, Research Policy, 29
(2): 313-330.
faber, J.a. 1993, De Geschiedenis van de Landbouwuniversiteit Wageningen, Deel III ‘Van Revolutie naar
Rendement, 1970-1990’, landbouwuniversiteit Wageningen, Wageningen.
fisher, d., atkinson-grosjean, J. & house, d. 2001, changes in academy/industry/state relations in
canada: The creation and development of the networks of centres of excellence, Minerva, 39
(3): 299-325.
freeman, c. 1997, “The diversity of national research systems” in Science in Tommorow’s Europe, eds. r.
Barre, M. gibbons, J. Maddox, B.r. Martin & P. Papon, economica international, Paris, p. 5-32.
funtowicz, S. & ravetz, J. 1993, Science for the post-normal age, Futures, 25: 735-755.
geiger, r.l. & Sa, c.M. 2008, Tapping the Riches of Science: Universities and the Promise of Economic
Growth, harvard university Press, cambridge, Massachusetts.
geuna, a. & Martin, B.r. 2003, university research evaluation and funding: an international compari-
son, Minerva, 41: 277-304.
geuna, a. & nesta, l.J.J. 2006, university patenting and its effects on academic research: The emerging
european evidence, Research Policy, 35: 790-807.
gibbons, M. 1999, Science’s new social contract with society, Nature, 402: c81-c84.
gibbons, M., limoges, c., nowotny, h., Schwartzman, S., Scott, P. & trow, M. 1994, The New Produc-
tion of Knowledge: The Dynamics of Science and Research in Contemporary Societies, Sage, london.
giddens, a. 1984, The Constitution of Society: Outline of the Theory of Structuration, Polity Press, cam-
bridge, england.
glänzel, W. 2006, on the h-index: a mathematical approach to a new measure of publication activity
and citation impact, Scientometrics, 67 (2): 315-321.
gläser, J., lange, S., laudel, g. & Schimank, u. 2010, “The limits of universality: how field-specific
epistemic conditions affect authority relations and their consequences” in Reconfiguring Knowledge
Production: Changing Authority Relationships in the Sciences and their Consequences for Intellectual
Innovation, eds. r. Whitley, J. gläser & l. engwall, oxford university Press, oxford.
godin, B. 1998, Writing performative history: The new new atlantis? Social Studies of Science, 28 (3):
465-483.
211
references
godin, B. & gingras, Y. 2000, The place of universities in the system of knowledge production, Research
Policy, 29 (2): 273-278.
göktepe-hulten, d. & Mahagaonkar, P. forthcoming, inventing and patenting activities of scientists: in
the expectation of money or reputation? Journal of Technology Transfer, doi: 10.1007/s10961-
009-9126-2.
goldfarb, B. 2008, The effect of government contracting on academic research: does the source of fun-
ding affect scientific output? Research Policy, 37 (1): 41-58.
griesemer, J.r. & gerson, e.M. 1993, collaboration in the Museum of vertebrate Zoology, Journal of the
History of Biology, 26 (2): 185-203.
grin, J., felix, f., Bos, B. & Spoelstra, S. 2004, Practices for reflexive design: lessons from a dutch pro-
gramme on sustainable agriculture, International Journal of Foresight and Innovation Policy, 1 (1):
126-149.
grin, J., van de graaf, h. & vergragt, P.J. 2003, een derde generatie milieubeleid: een sociologische
perspectief en een beleidswetenschappelijk programma, Beleidswetenschap, 17 (1): 51-72.
grin, J. 2010, “Part iii: understanding transitions from a governance perspective” in Transitions to Sus-
tainable Development: New Directions in the Study of Long Term Transformative Change, eds. J. grin,
J. rotmans & J.W. Schot, routledge, , p. 221-319.
groenewegen, P. 1988, Scientists, Audiences and Resources: The Development of Dutch Toxicological Re-
search, Phd thesis, university of amsterdam.
guston, d.h. 2000, Between Politics and Science: Assuring the Integrity and Productivity of Research, cam-
bridge university Press, cambridge.
guston, d.h. & Kenniston, K. 1994a, The Fragile Contract: University Science and the Federal Government,
Mit Press, cambridge, Massachusetts.
guston, d.h. & Kenniston, K. 1994b, “introduction: the social contract for science” in The Fragile Con-
tract: University Science and the Federal Government, eds. d.h. guston & K. Kenniston, Mit Press,
cambridge, Massachusetts, p. 1-41.
hackett, e.J. 2005, introduction to the special guest-edited issue on scientific collaboration, Social Stu-
dies of Science, 35: 667-671.
hackett, e.J., conz, d., Parker, J., Bashford, J. & delay, S. 2004, tokamaks and turbulence: research
ensembles, policy and technoscientific work, Research Policy, 33: 747-767.
halffman, W. 2003, Boundaries of Regulatory Science: Eco/Toxicology and Aquatic Hazards of Chemicals in
the US, England, and the Netherlands, 1970-1995, Phd thesis, university of amsterdam.
hayes, a.W. 2007, Principles and Methods of Toxicology, informa healthcare, new York.
212
references
heimeriks, g., van den Besselaar, P. & frenken, K. 2008, digital disciplinary differences: an anlysis of
computer-mediated science and ‘Mode 2’ knowledge production, Research Policy, 37 (9): 1602-
1615.
hemlin, S. & rasmussen, S.B. 2006, The shift in academic quality control, Science Technology and Human
Values, 31 (2): 173-198.
hessels, l.K., de Jong, S. & van lente, h. 2010, “Multidisciplinary collaborations in toxicology and
paleo-ecology: equal means to different ends” in Collaboration in the New Life Sciences, eds. J.n.
Parker, n. vermeulen & B. Penders, ashgate, london, p. 37-62.
hessels, l.K. & van lente, h. 2010a, The mixed blessing of Mode 2 knowledge production, Science,
Technology & Innovation Studies, 6 (1): 65-69.
hessels, l.K. & van lente, h. 2010b, reward research that benefits society, with kudos or even cash,
Nature, 465: 685.
hessels, l.K. & van lente, h. 2008, re-thinking new knowledge production: a literature review and a
research agenda, Research Policy, 37: 740-760.
hessels, l.K., van lente, h. & Smits, r.e.h.M. 2009, in search of relevance: the changing contract
between science and society, Science and Public Policy, 36 (5): 387-401.
hicks, d. 2009, evolving regimes of multi-university research evaluation, Higher Education, 57: 393-404.
hicks, d.M. & Katz, J.S. 1996, Where is science going? Science Technology and Human Values, 21 (4):
379-406.
hoes, a., regeer, B.J. & Bunders, J.f.g. 2008, transformers in knowledge production: Building science-
practice collaborations, Action Learning: Research and Practice, 5 (3): 207-220.
homburg, e. 2003, Speuren op de tast: Een historische kijk op industriële en universitaire research, universi-
teit Maastricht, Maastricht.
homburg, e. & Palm, l. (eds.) 2004a, De Geschiedenis van de Scheikunde in Nederland 3, delft university
Press, delft.
homburg, e. & Palm, l. 2004b, “grenzen aan de groei - groei aan de grenzen: enkele ontwikkelings-
lijnen van de na-oorlogse chemie” in De Geschiedenis van de Scheikunde in Nederland 3, eds. e.
homburg & l. Palm, delft university Press, delft, p. 3-18.
huff, a.S. 2000, changes in organizational knowledge production, Academy of Management Review, 25
(2): 288-293.
huffman, W. & Just, r.e. 1999, The organization of agricultural research in western developed countries,
Agricultural Economics, 21: 1-18.
hutter, W. 2004, “chemie, chemici en wetenschapsbeleid” in De Geschiedenis van de Scheikunde in Neder-
land 3, eds. e. homburg & l. Palm, delft university Press, delft, p. 19-36.
213
references
irvine, J. & Martin, B.r. 1984, Foresight in Science: Picking the Winners, frances Pinter, london.
Jacob, M. 2009, on commodification and the governance of academic research, Minerva, 47: 391-405.
Jacob, M. 2000, “’Mode 2’ in context: the contract researcher, the university and the knowledge society”
in The Future of Knowledge Production in the Academy, eds. M. Jacob & t. hellström, Srhe and
open university Press, Buckingham, p. 11-27.
Jain, S., george, g. & Maltarich, M. 2009, academics or entrepreneurs? investigating role identity modi-
fication of university scientists involved in commercial activitiy, Research Policy, 38: 922-935.
Jansen, J.d. 2002, Mode 2 knowledge and institutional life: taking gibbons on a walk through a South
african university, Higher Education, 43 (4): 507-521.
Jantsch, e. 1972, Technological Planning and Social Futures, cassell, london.
Jasanoff, S. 2005, Designs on Nature: Science and Democracy in Europe and the United States, Princeton
university Press, Princeton.
Jongbloed, B. & Salerno, c. 2003, De Bekostiging van het Universitaire Onderwijs en Onderzoek in Ne-
derland: Modellen, Thema’s en Trends, center for higher education Policy Studies, universiteit
twente, enschede.
Katz, J.S. & Martin, B.r. 1997, What is research collaboration? Research Policy, 26 (1): 1-18.
Kaufmann, a. & Kasztler, a. 2009, differences in publication and dissemination practices between
disciplinary and transdisciplinary science and the consequences for research evaluation, Science
and Public Policy, 36 (3): 215-227.
Kaufmann, a. & tödtling, f. 2001, Science-industry interaction in the process of innovation: The im-
portance of boundary-crossing between systems, Research Policy, 30 (5): 791-804.
Kersten, a.e. 1996, Een organisatie van en voor onderzoekers: ZWO 1947-1988, van gorcum, assen.
Klein, J.t. 1990, Interdisciplinarity: History, Theory, and Practice, Wayne State university Press, detroit.
Kleinman, d.l. & vallas, S.P. 2001, Science, capitalism, and the rise of the “knowledge worker”: The
changing structure of knowledge production in the united States, Theory and Society, 30 (4): 451-
492.
Klerkx, l. & leeuwis, c. 2009, establishment and embedding of innovation brokers at different innova-
tion system levels: insights from the dutch agricultural sector, Technological Forecasting and Social
Change, 76: 849-860.
Klerkx, l. & leeuwis, c. 2008a, institutionalizing end-user demand steering in agricultural r&d: far-
mer levy funding of r&d in The netherlands, Research Policy, 37: 460-472.
Klerkx, l. & leeuwis, c. 2008b, delegation of authority in research funding to networks: experiences
with a multiple goal boundary organization, Science and Public Policy, 35 (3): 183-196.
214
references
Kloprogge, P. & van der Sluijs, J.P. 2006, The inclusion of stakeholder knowledge and perspectives in
integrated assessment of climate change, Climate Change, 75: 359-389.
Kncv & vnci 1994, Toekomstig chemisch onderzoek: Universitair fundament voor industriële meerwaar-
de.
Knorr-cetina, K.d. 1999, Epistemic Cultures, harvard university Press.
Knorr-cetina, K.d. 1982, Scientific communities or transepistemic arenas of research? a critique of
quasi-economic models of science, Social Studies of Science, 12: 101-130.
Kronjee, g. & nooteboom, B. 2008, “research, higher education, and innovation” in Micro-foundations
for Innovation Policy, eds. B. nooteboom & e. Stam, amsterdam university Press, amsterdam, p.
103-134.
lamont, M. 2009, How Professors Think: Inside the Curious World of Academic Judgment, harvard univer-
sity Press, cambridge, Ma.
latour, B. 1987, Science in Action, harvard university Press, cambridge, Massachusetts.
latour, B. & Woolgar, S. 1986, Laboratory Life: The Construction of Scientific Facts, Sage, london.
laudel, g. 2001, collaboration, creativity and rewards: why and how scientists collaborate, International
Journal of Technology Management, 22 (7): 762-781.
law, J. 1992, notes on the Theory of the actor-network: ordering, Strategy, and heterogeneity, Systems
Practice, 4 (5): 379-393.
lawrence, r.J. & despres, c. 2004, futures of transdisciplinarity, Futures, 36: 397-405.
lee, S. & Bozeman, B. 2005, The impact of research collaboration on scientific productivity, Social Stu-
dies of Science, 35 (5): 673-702.
leeuwis, c. 2000, learning to be sustainable: does the dutch agrarian knowledge market fail? The
Journal of Agricultural Education and Extension, 7 (2): 79-92.
leeuwis, c., Smits, r., grin, J., Klerkx, l., Mierlo, B.v. & Kuipers, a. 2006, “equivocations on the post-
privatization dynamics in agricultural innovation systems” in The Design of an Innovation-Enhan-
cing Environment, ed. r. Smits, transforum, Zoetermeer, p. 3-58.
leišytė, l. 2007, University Governance and Academic Research: Case Studies of Research Units in Dutch
and English Universities, Phd thesis, university of twente, enschede.
leišytė, l., enders, J. & de Boer, h. 2008, The freedom to set research agendas: illusion and reality of
the research units in the dutch universities, Higher Education Policy, 21: 377-391.
lenhard, J., lücking, h. & Schwechheimer, h. 2006, expert knowledge, Mode-2 and scientific discipli-
nes: two contrasting views, Science and Public Policy, 33 (5): 341-350.
215
references
lepori, B., van den Besselaar, P., dinges, M., Potì, B., reale, e., Slipsaeter, S., Thèves, J. & van der
Meulen, B. 2007, comparing the evolution of national research policies: what patterns of change?
Science and Public Policy, 34 (6): 372-388.
levidow, l., Søgaard, v. & carr, S. 2002, agricultural public-sector research establishments in Western
europe: research priorities in conflict, Science and Public Policy, 29 (4): 287-295.
leydesdorff, l. 2008, caveats for the use of citation analysis in research and journal evaluation, Journal
of the American Society for Information Science and Technology, 59 (2): 278-287.
leydesdorff, l. & Meyer, M. 2006, triple helix indicators of knowledge-based innovation systems:
introduction to the special issue, Research Policy, 35 (10): 1441-1449.
loging, W., harland, l. & Williams-Jones, B. 2007, high-throughput electronic biology: mining infor-
mation for drug discovery, Nature Reviews Drug Discovery, 6.
lorenz, c. 2008, If you’re so smart, why aren’t you rich? Universiteit, markt en management, Boom, amster-
dam.
lundvall, B.a. 1992, National Systems of Innovation: Towards a Theory of Innovation and Interactive Lear-
ning, Pinter, london.
lyall, c., Bruce, a., firn, J., firn, M. & tait, J. 2004, assessing end-use relevance of public sector research
organisations, Research Policy, 33 (1): 73-87.
Maat, h. 2003, het innovatiesysteem voor de nederlandse landbouw, NEHA-jaarboek, 66: 234-262.
Maat, h. 2001, Science Cultivating Practice: A History of Agricultural Science in the Netherlands and its
Colonies, 1863-1986, Kluwer, dordrecht.
Magner, l.n. 1994, A History of the Life Sciences, dekker, new York.
Martin, B.r. 2003, “The changing social contract for science and the evolution of the university” in Sci-
ence and Innovation: Rethinking the Rationales for Funding and Governance, eds. a. geuna, a. Salter
& W.e. Steinmueller, edward elgar, cheltenham, p. 7-29.
Meeus, M., oerlemans, l.a.g. & hage, J. 2004, industry-public knowledge infrastructure interaction:
intra- and inter-organizational explanations of interactive learning, Industry and Innovation, 11
(4): 327-352.
Melin, g. 2000, Pragmatism and self-organization. research collaboration on the individual level,
Research Policy, 29: 31-40.
Merkx, f. & van den Besselaar, P. 2008, Positioning indicators for cross-disciplinary challenges: the
dutch coastal defense research case, Research Evaluation, 17 (1): 4-16.
Meyer-Krahmer, f. & Schmoch, u. 1998, Science-based technologies: university-industry interactions
in four fields, Research Policy, 27 (8): 835-851.
216
references
Moed, h.f. & hesselink, f.t. 1996, The publication output and impact of academic chemistry research
in the netherlands during the 1980s: bibliometric analyses and policy implications, Research
Policy, 25: 819-836.
Morillo, f., Bordons, M. & gomez, i. 2003, interdisciplarity in science: a tentative typology of discipli-
nes and research areas, Journal of the American Society for Information Science and Technology, 54
(13): 1237-1249.
Morris, n. 2003, academic researchers as ‘agents’ of science policy, Science and Public Policy, 30 (5):
359-370.
Morris, n. 2000, Science policy in action: Policy and the researcher, Minerva, 38 (4): 425-451.
Morris, n. & rip, a. 2006, Scientists’ coping strategies in an evolving research system: the case of life
scientists in the uK, Science and Public Policy, 33 (4): 253-263.
Motohashi, K. 2005, university–industry collaborations in Japan: The role of new technology-based
firms in transforming the national innovation System, Research Policy, 34 (5): 583-594.
nelson, r. 1993, National Innovation Systems: A Comparative Analysis, oxford university Press, oxford.
nowotny, h., Scott, P. & gibbons, M. 2001, Re-Thinking Science: Knowledge and the Public in an Age of
Uncertainty, Polity Press, cambridge.
ocv 1995, Chemie in Perspectief: een verkenning van vraag en aanbod in het chemisch onderzoek, overleg-
commissie verkenningen, amsterdam.
oliver, c. 1991, Strategic responses to institutional processes, Academy of Management Review, 16 (1):
145-179.
Packer, K. & Webster, a. 1996, Patenting culture in science: reinventing the scientific wheel of credibi-
lity, Science, Technology & Human Values, 21 (4): 427-453.
Penders, B., horstman, K. & vos, r. 2009, large-scale research and the goal of health: doable problem
construction in ‘new’ nutrition science, Interdisciplinary Science Reviews, 34 (4): 327-344.
Pestre, d. 2003, regimes of knowledge production in society: towards a more political and social rea-
ding, Minerva, 41: 245-261.
Pfeffer, J. & Salancik, g.r. 1978, The External Control of Organizations: A Resource Dependence Perspec-
tive, harper & row, new York.
Pickstone, J.v. 2007, Working knowledge before and after circa 1800, Isis, 98: 489-516.
Pohl, c. 2005, transdisciplinary collaboration in environmental research, Futures, 37: 1159-1178.
Porter, a.l. & rafols, i. 2009, is science becoming more interdisciplinary? Measuring and mapping six
research fields over time, Scientometrics, 81 (3): 719-745.
Potì, B. & reale, e. 2007, changing allocation models for public research funding: an empirical explora-
tion based on project funding data, Science and Public Policy, 34 (6): 417-430.
217
references
reale, e. & Seeber, M. forthcoming, organisation response to institutional pressures in higher educa-
tion: the important role of the disciplines, Higher Education, doi: 10.1007/s10734-010-9322-x.
regeer, B.J. 2010, Making the Invisible Visible, Phd thesis, vrije universiteit, amsterdam.
regeer, B.J. & Bunders, J.f.g. 2003, The epistemology of transdisciplinary research: from knowledge
integration to communities of practice, Interdisciplinary Environmental Review, 5 (2): 98-118.
rip, a. 2007, a changing social contract between science and society, presented at ‘Science and its Publics’
conference in Münich, 24-25 June.
rip, a. 2004, Strategic research, post-modern universities and research training, Higher Education Policy,
17 (2): 153-166.
rip, a. 2002a, regional innovation systems and the advent of strategic science, Journal of Technology
Transfer, 27: 123-131.
rip, a. 2002b, “Science for the 21st century” in The Future of Science and the Humanities, eds. P. tinde-
mans, a. verrijn-Stuart & r. visser, amsterdam university Press, amsterdam, p. 99-148.
rip, a. 2000, “fashions, lock-ins and the heterogeneity of knowledge production” in The Future of Know-
ledge Production in the Academy, eds. M. Jacob & t. hellström, Srhe and open university Press,
Buckingham, p. 28-39.
rip, a. 1997, a cognitive approach to relevance of science, Social Science Information, 36 (4): 615-640.
rip, a. 1994, The republic of science in the 1990s, Higher Education, 28: 3-23.
rip, a. 1989, “Sturing en stuurbaarheid van de wetenschappen. het model van de Starnbergers” in We-
tenschapsleer, ed. M. Korthals, Boom/open universiteit, Meppel/heerlen, p. 187-206.
rip, a. 1988, “contextual transformations in contemporary science” in Keeping Science Straight: A
Critical Look at the Assessment of Science and Technology, ed. a. Jamison, department of Theory of
Science, university of gothenburg, gothenburg, p. 59-87.
rip, a. 1982, de gans met de gouden eieren en andere maatschappelijke legitimaties van de moderne
wetenschap, De Gids, 145 (5): 285-297.
rip, a. & Boeker, e. 1975, Scientists and social responsibility in the netherlands, Social Studies of Sci-
ence, 5: 457-484.
rip, a. & nederhof, a.J. 1985, Stimulering van de biotechnologie in Nederland: effecten op het wetenschap-
pelijke veld, distributiecentrum overheidspublikaties, den haag.
rip, a. & van der Meulen, B.J.r. 1996, The post-modern research system, Science and Public Policy, 23
(6): 343-352.
roberts, M.r. 2009, realizing societal benefit from academic research: analysis of the national Science
foundation’s broader impacts criterion, Social Epistemology, 23 (3): 199.
218
references
roseboom, J. & rutten, h. 1998, The transformation of the dutch agricultural research system: an unfi-
nished agenda, World Development, 26 (6): 1113-1126.
Sagotsky, J.a., Zhang, l., Wang, Z., Martin, S. & deisboeck, t. 2008, life Sciences and the web: a new
era for collaboration, Molecular Systems Biology, 4:201.
Salomon, J. 1980, “Wetenschapsbeleidstudies en de ontwikkeling van wetenschapsbeleid” in Macht over
Kennis, eds. a. rip & P. groenewegen, Samsom, alphen aan den rijn, p. 23-52.
Salter, a.J. & Martin, B.r. 2001, The economic benefits of publicly funded basic research: a critical
review, Research Policy, 30 (3): 509-532.
Schäfer, W. 1983, “normative finalization” in Finalization in Science, eds. g. Böhme, W. van den daele,
r. hohlfeld, W. Krohn & W. Schäfer, d. reidel, dordrecht, p. 207-231.
Schmoch, u. & Schubert, t. forthcoming, Strategic steering of research by new Public Management in
german universities: a looming state-science conflict? Research Evaluation (accepted for publica-
tion).
Schot, J.W. & van lente, d. 2003, “techniek, industrialisatie en de betwiste modernisering van neder-
land” in Techniek in Nederland in de Twintigste Eeuw: Part VII, Techniek en Modernisering: Balans
van de Twintigste Eeuw, eds. J.W. Schot, a. rip & h.W. lintsen, Walburg Pers, Zutphen.
Scott, a. 2007, Peer review and the relevance of science, Futures, 39: 827-845.
Shapley, d. & roy, r. 1985, Lost at the Frontier, iSi Press, Philadelphia.
Shinn, t. 2002, The triple helix and new production of knowledge: Prepackaged thinking on science
and technology, Social Studies of Science, 32 (4): 599-614.
Shove, e. 2003, Principals, agents and research programmes, Science and Public Policy, 30 (5): 371-381.
Shrum, W., genuth, J. & chompalov, i. 2007, Structures of Scientific Collaboration, Mit Press, cam-
bridge, Ma.
Slaughter, S. & leslie, l.l. 1997, Academic Capitalism: Politics, Policies, and the Entrepreneurial University,
The John hopkins university Press, Baltimore.
Small, h., Kushmerick, a. & Benson, d. 2008, Scientists’ perceptions of the social and political implica-
tions of their research, Scientometrics, 74 (2): 207-221.
Smits, r. 2002, innovation studies in the 21st century: Questions from a user’s perspective, Technological
Forecasting and Social Change, 69: 861-883.
Smits, r. & Kuhlmann, S. 2004, The rise of systemic instruments in innovation policy, International
Journal of Foresight and Innovation Policy, 1 (1): 4-32.
Spaapen, J. 1995, Research for Society: Towards a method for the assessment of research in a policy context,
university of amsterdam, amsterdam.
219
references
Spaapen, J., dijstelbloem, h. & Wamelink, f. 2007, Evaluating Research in Context, coS, The nether-
lands, The hague.
Spiegel-rösing, i. 1980, “Wetenschapssturing” in Macht over Kennis, eds. a. rip & P. groenewegen,
Samsom, alphen aan den rijn, p. 53-76.
Starkey, K. & Madan, P. 2001, Bridging the relevance gap: aligning Stakeholders in the future of Ma-
nagement research, British Journal of Management, 12 (s1): s3-s26.
Steele, c., Butler, l. & Kingsley, d. 2006, The publishing imperative: the pervasive influence of publica-
tion metrics, Learned Publishing, 19 (4): 277-290.
Stengers, i. 1997, Power and Invention: Situating Science, university of Minnesota Press, Minneapolis.
Stokes, d.e. 1997, Pasteur’s Quadrant: Basic Science and Technological Innovation, Brookings institution
Press.
technopolis & dialogic 2007, Evaluatie Vernieuwingsimpuls 2000-2006, dialogic, utrecht.
Thakur, n.l., Jain, r., natalio, f., hamer, B., Thakur, a.n. & Müller, W.e.g. 2008, Marine molecular
biology: an emerging field of biological sciences, Biotechnology Advances, 26 (3): 233-245.
tindemans, P. & Soete, l. 2007, Assessment of the feasibility and possible impact of the establishment of a
European Institute of Technology, Study for the european Parliament, Brussels.
tijssen, r.J.W. & Korevaar, J.c. 1997, unravelling the cognitive and interorganisational structure of pu-
blic / private r&d networks: a case study of catalysis research in the netherlands, Research Policy,
25: 1277-1293.
tinoco, i. 1987, Report on the Human Genome Initiative, Subcommittee of the health and environmental
research advisory committee (heRAc).
tuunainen, J. 2005, hybrid practices? contributions to the debate on the mutation of science and uni-
versity, Higher Education, 50 (2): 275-298.
van den daele, W. 1978, “The ambivalent legitimacy of the pursuit of knowledge” in Science, Society and
Education, eds. e. Boeker & M. gibbons, vu Bookshop, amsterdam.
van der Meulen, B. 2008, “interfering governance and emerging centres of control” in The Changing
Governance of the Sciences: The Advent of Research Evaluation Systems, eds. r. Whitley & J. gläser,
Springer, dordrecht, p. 191-204.
van der Meulen, B. & rip, a. 1998, Mediation in the dutch science system, Research Policy, 27: 757-769.
van der Meulen, B.J.r. 1998, Science policies as principal-agent games: institutionalization and path
dependency in the relation between government and science, Research Policy, 27: 397-414.
van der Meulen, B.J.r. 1992, Beoordelingsprocessen in Wetenschap, proefschrift universiteit twente,
enschede.
220
references
van der Meulen, B.J.r. & rip, a. 2000, evaluation of societal quality of public sector research in the
netherlands, Research Evaluation, 8 (1): 11-25.
van der Ploeg, J.d. & ettema, M. (eds.) 1990, Tussen Bulk en Kwaliteit: De Struktuur van de Voedselpro-
duktieketen, van gorcum, assen.
van der Sluijs, J.P., craye, M., funtowicz, S., Kloprogge, P., ravetz, J. & risbey, J. 2005, combining
quantitative and qualitative measures of uncertainty in model-based environmental assessment:
The nuSaP system, Risk Analysis, 25 (2): 481-492.
van helvoort, t. 2005, De KNAW tussen wetenschap en politiek: De positie van de scheikunde in de Akade-
mie in naoorlogs Nederland, KnaW, amsterdam.
van hemert, P., nijkamp, P. & verbraak, J. 2009, evaluating social science and humanities knowledge
production: an exploratory analysis of dynamics in science systems, Innovation: The European
Journal of Social Science Research, 22 (4): 443.
van lente, h. 2006, “Prospective structures of science and science policy” in Innovation, Science, and
Institutional Change, eds. J. hage & M. Meeus, oxford university Press, oxford.
van lente, h. 1993, Promising Technology: The Dynamics of Expectations in Technological Developments,
eburon, delft.
van lente, h. & rip, a. 1998, The rise of membrane technology: from rhetorics to social reality, Social
Studies of Science, 28 (2): 221-254.
van lente, h. & van til, J.i. 2008, articulation of sustainability in the emerging field of nanocoatings,
Journal of Cleaner Production, 16 (8-9): 967-976.
van rijnsoever, f.J., hessels, l.K. & vandeberg, r.l.J. 2008, a resource-based view on the interactions
of university researchers, Research Policy, 37: 1255-1266.
vasileiadou, e. 2009, Working Apart Together: Using ICTs in research collaboration, Phd thesis, university
of amsterdam, amsterdam.
vavakova, B. 1998, The new social contract between governments, universities and society: has the old
one failed? Minerva, 36: 209-228.
vermeulen, n. 2009, Supersizing Science: On building large-scale research projects in Biology, universitaire
Pers Maastricht, Maastricht.
versleijen, a. (ed.) 2007, Dertig jaar publieke onderzoeksfinanciering in Nederland 1975-2005: Historische
trends, actuele discussies, rathenau instituut, den haag.
vSnu, nWo & KnaW 2003, Standard Evaluation Protocol 2003-2009 for Public Research Organisations.
Wamelink, f. & Spaapen, J. 1999, De evaluatie van universitair onderzoek: methodiek voor het incorporeren
van de maatschappelijke waarde van onderzoek (No. NRLO 99/12), nationale raad voor land-
bouwkundig onderzoek, den haag.
221
references
Weingart, P. 2005, impact of bibliometrics upon the science system: inadvertent consequences? Sciento-
metrics, 62 (1): 117-131.
Weingart, P. 1997, from “finalization” to “Mode 2”: old wine in new bottles? Social Science Information,
36 (4): 591-613.
Whitley, r. 2000, The Intellectual and Social Organization of the Sciences, oxford university Press, oxford.
Whitley, r. & gläser, J. (eds.) 2007, The Changing Governance of the Sciences, Springer, dordrecht.
Wissema, J.g. 2009, Towards the Third Generation University: Managing the University in Transition,
edward elgar, cheltenham.
Wouters, P. 1999, The Citation Culture, Phd thesis, university of amsterdam, amsterdam.
Wouters, P. 1997, citation cycles and peer review cycles, Scientometrics, 38 (1): 39-55.
Yearley, S. 2005, Making Sense of Science: Understanding the Social Study of Science, Sage, london.
Yin, r.K. 2003, Case Study Research: Design and Methods, Sage, Thousand oaks, california.
Ziman, J. 2003, non-instrumental roles of science, Science and Engineering Ethics, 9 (1): 17-27.
Ziman, J. 2000, Real Science: What it is, and what it means, cambridge university Press, cambridge.
Ziman, J. 1994, Prometheus Bound: Science in a Dynamic Steady State, cambridge university Press, cam-
bridge.
Zomer, a., enders, J. & Jongbloed, B. 2010, do spin-offs make the academics’ head spin? The impacts of
spin-off companies on their parent research organisation, Minerva, forthcoming.
223
Summary
The starting assumption of this thesis is that academic researchers struggle for relevance. to a lesser or
greater extent, they need to make sure that their work is valuable to society, either directly or indirectly.
This can give personal satisfaction, but it also helps to legitimate one’s work to the outside world and to
get access to resources. however, striving for relevance is not a straightforward exercise. first, resear-
chers may face difficulties when trying to reconcile the need for relevance with other plans, ambitions
and values. Second, the meaning of relevance is not fixed; it is the product of social interactions among
scientists, policy makers and other stakeholders. Because of these aspects it makes sense to speak of a
‘struggle’.
in this thesis, the struggle for relevance is defined as the combination of the efforts of scientists to
make their work correspond with ruling standards of relevance and their efforts to influence these stan-
dards of relevance. With this study of the relevance of science i aim to contribute to the understanding
of transformations in the knowledge infrastructure, as discussed in a large and expanding literature using
concepts like Mode 2 knowledge production, post-academic science and the triple helix of university-
government-industry relationships.
The main research question of this thesis is: how to understand the changes in the struggle for
relevance of dutch academic researchers in chemistry, biology and agricultural science, in the period
1975-2005?
My theoretical starting point is a distinction between the agency of scientists and the structures
patterning their behavior. in line with structuration theory, i assume that these structures are both
medium and outcome of research practices. further i regard science as a social activity, and the progress
of science as the outcome of socially embedded actions of researchers. another starting assumption is
that scientists are involved in a continuous struggle for resources, as they depend on their environment
for funding and legitimacy. Based on this theoretical foundation, in this thesis i develop and use three
central concepts to analyze struggles for relevance: the credibility cycle, the science-society contract,
and the research system.
i use the credibility cycle to conceptualize the struggles for relevance of individual scientists. This
model explains how struggles for reputation influence the daily work of individual scientists and depicts
224
Summary
the research process as a repetitive cycle in which conversions take place between money, staff, data,
arguments, articles, recognition, and so on. i investigate the changing structures influencing scientific
agency from two different angles, using the concepts of a science-society contract, and a research system,
respectively. The contract refers to all implicit and explicit agreements between science and societal par-
ties about what science should do, why it should do this, and what are the appropriate conditions for sci-
ence to function well. The research system draws attention to the institutional environment of academic
research, consisting of research councils, firms, public research institutes, governmental departments,
and other organizations that provide incentives and constraints to conduct particular kinds of research.
This research follows a case study approach and deals with chemistry, biology and agricultural sci-
ence in the netherlands. Within each case i focus in particular on two or three fields representing the
breadth of the discipline in terms of possible stakeholders. This thesis deals primarily with academic
knowledge production, that is, research conducted at universities. data have been drawn from in-depth
interviews with 47 academic researchers, and with a number scholarly experts and representatives of
various organizations in the research system. Moreover i have analyzed a selection of governmental
policy documents, reports and strategic plans of research councils, foresight studies, evaluations and
other important publications about the disciplines.
after an introductory chapter presenting the background of this study, its theoretical framework and
methodology, chapter 2 provides a state-of-the-art of the current literature on the present debate on
transformations in the knowledge infrastructure. The notion of ‘Mode 2 knowledge production’ (gib-
bons et al, 1994) was chosen as a starting point in this literature review, because of its prominence in this
debate, both in academia and policy circles. i systematically review its reception in scientific literature
and compare this concept to a number alternative diagnoses of changing science systems, such as the
‘triple helix’ and ‘Post-normal science’. i conclude that the strength of The new Production of Knowl-
edge (nPK) is its exceptionally wide scope, as it deals with changes on the cognitive, organizational and
societal level. Thanks to its breadth, nPK has created a forum to discuss a wide range of putative trends
in the science system. however, based on the review of critical reactions in scholarly literature i also
identify a number of problems, regarding the empirical validity, the conceptual strength and the political
value of nPK. The most important problems are the lack of empirical support and the questionable
coherence of the five constitutive features of ‘Mode 2’. for this reason i recommend separating these
features and investigating each separately. The review is concluded with a research agenda for scholars in
Science, technology and innovation Studies, dealing with the three most controversial features of Mode
2 knowledge production: transdisciplinarity, reflexivity, and quality control. This agenda has formed the
starting point for formulating the research question of this thesis, as presented above.
after the literature review, my next step is a further conceptualization of (struggles for) relevance in
terms of my theoretical framework. chapter 3 develops the first two building blocks of this framework:
the science-society contract and the credibility cycle. Based on a brief review of the usage of the contract
225
Summary
metaphor in the literature, i propose to use it as a heuristic for studying the changing relationship
between academic science and society. i suggest regarding its content in terms of three main compo-
nents, dealing with the identity of science, the rationales for supporting it, and the conditions under
which university researchers work, respectively. The second element of my framework is the concept of
the credibility cycle. in a brief case study of dutch academic chemistry the usefulness of these concepts
is illustrated. in this case study i explore how changing concepts of ‘relevance’ in the science-society
contract influence scientific practices, with a particular focus on the institutions governing the credibility
cycle. This analysis gives a first indication of the tension between the increasing pressure for scientific
publications and the increasing pressure for relevant outcomes – increasingly defined in terms of practi-
cal applications.
The following three chapters (chapters 4-6) use and enrich (elements of) this framework in the
analysis of case studies of dutch chemistry, biology, and agricultural science, respectively. each study ad-
dresses similar questions about the changing struggle for relevance. chapter 4 presents a more elaborate
study of the chemistry case. Between 1975 and 2005 four major changes can be identified in its science-
society contract:
1. the relative share of funding for application oriented university research has grown;
2. the interactions with industry have become more intensive;
3. universities attempt to play an active role in the valorization of research outcomes; and
4. systematic performance evaluations have become a powerful institution governing academic re-
search.
The consequences of these developments for the struggles for relevance of researchers vary across the
three chemical fields studied. in catalysis linking with industrial applications helps in many steps of the
credibility cycle. Practical applications yield much less credibility in environmental chemistry, where
application oriented research agendas help to acquire funding, but not to publish prestigious papers or
to earn peer recognition. in biochemistry practical applications hardly help in gaining credibility, as this
field is still strongly oriented at fundamental questions. The differences between the fields can be ex-
plained by the presence or absence of powerful upstream end-users that can afford to invest in academic
research that promises long term benefits.
chapter 5 addresses the influence of changes in the contract on the actual behavior of academic
researchers and the content of their work. it focuses in particular on multidisciplinary collaborations in
paleo-ecology and toxicology. This chapter shows that the societal contract for biology has undergone
a development similar to the one of chemistry. The identity of academic biology has changed from a
basic to a strategic science; rationales for funding biology increasingly emphasize practical applications;
the available funding has shifted and performance evaluations have emerged. These developments turn
out to provide both pressures and incentives for multidisciplinarity that differ between the two fields
under study. in paleo-ecology a rise in multidisciplinary research collaborations can be understood as a
226
Summary
response to outside pressures to do more ‘relevant’ research, particularly in the area of climate change. in
toxicology i observe the opposite: multidisciplinarity is used to strengthen the fundamental research on
physiological mechanisms of toxicity. i conclude that multidisciplinary research collaborations can serve
various ends, ranging from enhancing the practical value of a fundamental field to strengthening the
fundamental basis of an applied field.
The central question of chapter 6 is: how do changing institutions influence academic research prac-
tices in dutch agricultural sciences? This chapter deals in particular with three fields of animal science.
These fields have experienced a highly dynamic societal context. dutch agricultural sciences are under
pressure to change the content and the organization of the research in order to contribute to more com-
petitive and sustainable agricultural production. in this chapter i analyze the changes in the institutions
of the agricultural research system and their consequences on the daily work of dutch animal scientists.
in the period studied, i observe a shift from the consensus-driven ‘ovo triptych’ to a heterogeneous
knowledge system. This shift entails two major changes in the institutional environment of academic
agricultural research: shifts in the available funding, and the rise of systematic performance evaluations.
as a result, in two fields the interactions with stakeholders have increased, but for different reasons. in
animal Breeding & genetics (aBg) researchers have frequent contact with breeding firms in order to
select challenging research issues, to collect data, and to acquire research funding. Thanks to these in-
teractions, aBg manages to obtain sufficient external funding, without compromising its basic research
agenda. in animal Production Systems (aPS) contacts with farmers only contribute to data collection,
while contacts with applied research institutes and with ngos help to get access to funding. This seems
a vulnerable combination, as aPS struggles to build a coherent research agenda and to publish sufficient
scientific papers. in cell Biology, the third field studied, interactions with ‘users’ have decreased rather
than increased, as there is a tendency to shift to more fundamental research issues with a higher potential
for prestigious publications, but which are not of immediate interest to the agricultural sector.
in chapter 7 i provide a further reflection on the material presented in the previous chapters. i
answer the general research question by subsequently addressing three sub-questions.
1. What changes can be discerned in the struggles for relevance of Dutch academic researchers in chemis-
try, biology and agricultural science, in the period 1975-2005?
across all scientific fields, i observed three common developments. first, in most fields the struggle for
relevance during the actual research process has intensified. in several fields interactions with possible
users of research outcomes were already common practice, but in general their frequency and salience
have grown. The second common trend relates to the way scientists earn recognition: on average, the
value of practical applications as a source of academic recognition has decreased. Since the 1970s
publication achievements have (further) pushed away social or economic impact as a source of academic
227
Summary
recognition. Third, the struggle for relevance in the context of funding acquisition has intensified. in all
fields the scientists i have interviewed report that aligning their work with the knowledge needs of socie-
tal stakeholders has become increasingly important for gathering sufficient funding. t
The changes in the struggle for relevance also varied across scientific fields. first, although the role of
stakeholders in the academic research process has generally grown, the extent to which they have become
involved in data collection varies strongly across fields. Second, there is variation in the (limited) degree
to which practical relevance is rewarded in terms of academic recognition. Third, in all fields promising
societal benefits can help to acquire funding, but the degree of involvement of societal stakeholders in the
actual agenda-setting differs. a fourth dimension concerns the tension that has arisen in some fields due
to the combination of these trends. in some fields there is a synergy between scientific productivity and
practical applications because interactions with stakeholders do not only help to acquire funding, but
they are also helpful in other credibility conversions. in other fields there is a tension between scientific
productivity and practical applications, as enhancing the industrial, medical or agricultural relevance of
one’s work implies a distantiation from these fields’ central debates. to this end, one would have to shift
to other model systems or to address other research questions which are less suitable for publishing in
prestigious journals and which are less likely to yield academic recognition.
2. How can the similarities among the changes in different scientific fields be explained?
The general trends across the fields studied can be understood as effects of the combination of three
major changes in the structural conditions of academic research in the netherlands. The first is the
diversification of funding sources. Between 1975 and 2005 the share of application oriented funding has
increased. Simultaneously, support for talented researchers has emerged, rewarding scientists who are
highly productive in terms of scientific publications. The second structural change that has shaped strug-
gles for relevance in all scientific fields studied is the rise of performance evaluations, which are domi-
nated by bibliometric quality indicators. The third important structural change concerns the dominant
perspective on the mission of the university in the netherlands. Since 1975 notions like accountability,
valorization and entrepreneurship have gradually replaced autonomy and independence as the ruling
values connected with academic research.
3. How can the differences among the changes in different scientific fields be explained?
one of the most striking findings of this thesis is that the changes in the struggle for relevance vary
strongly across scientific fields. a possible explanation for the differences among scientific fields can be
found in the combination of the degree of convergence in their search pattern, their traditional commu-
nication culture, and characteristics of their upstream end-users.
228
Summary
first, in divergent fields, which have a high strategic task uncertainty, it seems more likely that niches
develop which fit the knowledge needs of societal stakeholders than in convergent fields in which there
are higher penalties on deviating from the shared research agenda.
Second, the extent to which practical applications count as a source of credibility can be understood
when taking into account the traditional communication culture of a scientific field. in fields with a ‘ru-
ral’ communication style in which there is a low people-to-problem ratio, there is usually a lower citation
density. This implies that bibliometric quality indicators have limited validity, so that it is less likely that
recognition will be based on publications only and more likely that practical outcomes such as policy ad-
vice, patents or spin-off firms count as well. in more competitive, urban fields (high people-to-problem
ratio), with a high citation density, bibliometric quality indicators will be more readily used, not only in
formal evaluations and management decisions, but also in informal processes of exchanging recognition.
Third, a possible explanation for the variation in the relationship between practical applications and
scientific productivity can be found in characteristics of other actors in the research system, in particular
the end-users of academic research. The fields that turned out most successful in combining stakeholder
interactions with academic performance, were fields with wealthy and powerful upstream end-users with
a long-term vision on the utility scientific research. By sponsoring fundamental research activities, which
provide good opportunities for high-impact publications, chemical industry (in the case of catalysis) and
animal breeding industry (aBg) support scientists along the complete cycle of credibility.
My findings have two major implications for the debate about changing science systems. first, my
thesis shows that the increasing pressure for productivity, as measured in bibliometric terms, can coun-
teract the pressure for practical utility. in many fields i observed a tension between research activities
that are useful for societal stakeholders and activities that are most promising in terms of scientific pub-
lications. Second, my work indicates that a further differentiation of science (policy) studies is needed,
as the dynamics of science vary much more across scientific fields than most literature suggests. Both
findings challenge the empirical validity of some central claims around the notion of Mode 2 knowledge
production, in particular regarding transdisciplinarity, reflexivity, and novel quality control.
Based on my findings i have formulated two sets of recommendations for science and innovation policy.
regarding the funding of university research, i warn for a restricted (short-term economic) concept of
valorization. The emphasis on short-term benefits leads to ineffective funding instruments, overrates the
value of academic patents, endangers the viability of more fundamental research lines, and it threatens
the stability of academic research groups and the coherence of their activities. i recommend to:
• govern academic research based on expected long-term benefits rather than short-term benefits;
• loosen the ‘valorization’ requirements for grants from the national research council nWo;
• make a more careful cost-benefit analysis before patenting academic inventions; and
• restore and protect the size of unconditional block grant support for university research.
229
Summary
With regard to research evaluation, i point to the perverse side-effects of bibliometric quality indica-
tors. They restrict the concept of academic excellence, promote a monoculture of science and create
inflation of the value of a scientific paper. to overcome these problems, bibliometric indicators should
be complemented both with qualitative assessments and with indicators of social and economic impact.
i recommend to:
• avoid direct linkage of funding decisions with bibliometric indicators;
• broaden the criteria of performance evaluations and include indicators of societal impact when ap-
plicable; and
• broaden the criteria for hiring and promoting academic staff: take into account teaching qualities
and societal impact as well.
231
Samenvatting
het uitgangspunt van dit proefschrift is dat academische onderzoekers strijden om relevantie. in meer of
mindere mate proberen ze ervoor te zorgen dat hun werk waardevol is voor de samenleving, direct of
indirect. dit kan ze persoonlijke voldoening geven, maar het helpt ze ook om hun werk te legitimeren
naar de buitenwereld en om toegang te krijgen tot de benodigde middelen. het streven naar relevantie
is echter niet altijd gemakkelijk; het heeft vaak het karakter van een worsteling. ten eerste kan het lastig
zijn om het najagen van relevantie te verzoenen met andere plannen, ambities en waarden. ten tweede
heeft relevantie geen vaste betekenis; het is het product van sociale interacties tussen wetenschappers,
beleidsmakers en andere betrokkenen. Kortom, de strijd om relevantie is een interessant maar ingewik-
keld krachtenspel.
in dit proefschrift heb ik de strijd om relevantie gedefinieerd als het geheel van inspanningen van
wetenschappers om hun werk in overeenstemming te brengen met gangbare standaarden van relevantie
en hun inspanningen om die standaarden te beïnvloeden. het doel van dit onderzoek naar relevantie
is om bij te dragen aan het begrip van veranderingen in de kennisinfrastructuur, die de afgelopen 15
jaar met concepten als Mode 2 kennisproductie, post-academische wetenschap, en de triple helix van
universiteit-overheid-bedrijven relaties werden aangeduid.
De onderzoeksvraag van dit proefschrift luidt: hoe kunnen we veranderingen in de strijd om relevantie van
Nederlandse academische onderzoekers in de chemie, biologie en landbouwwetenschappen in de periode 1975-
2005 begrijpen?
Mijn theoretische vertrekpunt is het onderscheid tussen de agency van wetenschappers en de
structuren die hun gedrag vormgeven. in lijn met structuratietheorie neem ik aan dat deze structuren
tegelijk medium en uitkomst zijn van onderzoekspraktijken. verder vat ik wetenschap op als een sociale
activiteit, en de voortgang van wetenschap als de uitkomst van sociaal ingebedde handelingen van
onderzoekers. een andere basisaanname is dat wetenschappers betrokken zijn bij een voortdurende
strijd om hulpbronnen, omdat ze afhankelijk zijn van hun omgeving voor financiering en legitimatie. in
dit proefschrift ontwikkel en gebruik ik drie centrale concepten om de strijd om relevantie te analyseren:
de geloofwaardigheidcyclus (credibility cycle), het contract tussen wetenschap en samenleving, en het
onderzoekssysteem.
232
Samenvatting
ik gebruik de geloofwaardigheidcyclus om de strijd om relevantie van individuele wetenschappers te
conceptualiseren. dit model legt uit hoe het dagelijkse werk van individuele wetenschappers verbonden
is met het opbouwen van een reputatie. het beeldt het onderzoeksproces af als een repeterende cyclus
waarin omzettingen plaatsvinden tussen geld, personeel, data, argumenten, artikelen, erkenningen,
enzovoorts. de veranderende structuren die het gedrag van wetenschappers beïnvloeden, onderzoek ik
vanuit twee verschillende invalshoeken, gebruikmakend van de begrippen ‘contract’ (tussen wetenschap
en samenleving) en ‘onderzoekssysteem’. het contract verwijst naar alle impliciete en expliciete overeen-
komsten tussen wetenschap en maatschappelijke partijen over wat wetenschap moet doen, waarom het
dit zou moeten doen, en wat de geschikte omstandigheden zijn voor wetenschap om goed te function-
eren. het onderzoekssysteem vertegenwoordigt de institutionele omgeving van academisch onderzoek,
bestaande uit onderzoeksfondsen, bedrijven, publieke onderzoeksinstituten, ministeries en andere
organisaties die prikkels en beperkingen opleggen om een bepaald soort onderzoek uit te voeren.
dit onderzoek is een combinatie van casestudies naar scheikunde, biologie en landbouwweten-
schappen in nederland. Binnen elke case richt ik me in het bijzonder op twee of drie onderzoeksgebie-
den die samen de breedte van deze discipline vertegenwoordigen in termen van mogelijke belangheb-
benden. dit proefschrift gaat in de eerste plaats over academische kennisproductie, dus onderzoek dat
wordt uitgevoerd aan universiteiten. gegevens voor het onderzoek zijn verzameld door middel van
diepte-interviews met 47 universitaire onderzoekers, en met een aantal deskundigen en vertegenwoordi-
gers van verschillende organisaties in het onderzoekssysteem. daarnaast heb ik een verzameling beleids-
documenten, rapporten en strategische plannen van onderzoeksfondsen, verkenningen, evaluaties en
andere belangrijke publicaties over de drie disciplines geanalyseerd.
na een inleidend eerste hoofdstuk dat de achtergrond, het theoretische kader en de methodologie
van deze studie behandelt, biedt hoofdstuk 2 een overzicht van de recente literatuur over transforma-
ties in de kennisinfrastructuur. voor deze literatuurstudie heb ik het begrip ‘Mode 2 kennisproductie’
(gibbons et al., 1994), gekozen als vertrekpunt vanwege haar prominente positie in het debat, zowel
aan de universiteit als in beleidskringen. ik geef een systematisch overzicht van de ontvangst van The
new Production of Knowledge (nPK) in de wetenschappelijke literatuur en vergelijk het begrip Mode
2 met een aantal alternatieve diagnoses van veranderende wetenschapssystemen, zoals de ‘triple helix’
en ‘Post-normal science’. Mijn conclusie luidt dat de kracht van nPK haar uitzonderlijk grote reikwijdte
is: het gaat over veranderingen op het cognitieve, organisatie- en maatschappelijke niveau. dankzij deze
reikwijdte heeft nPK een forum geschapen voor het bespreken van verschillende vermoedelijke trends
in het wetenschapssysteem. op basis van een overzicht van de kritische reacties in de wetenschappeli-
jke literatuur, identificeer ik echter ook een aantal problemen aangaande de empirische validiteit, de
conceptuele kwaliteit en de politieke kleur van nPK. de belangrijkste problemen zijn het gebrek aan
empirische ondersteuning en de twijfelachtige samenhang van de vijf essentiële kenmerken van ‘Mode
2’. daarom raad ik aan om deze kenmerken van elkaar te scheiden en ieder apart te onderzoeken. het
233
Samenvatting
literatuuroverzicht wordt afgesloten met een onderzoeksagenda voor onderzoekers op het gebied van
Wetenschaps-, technologie- en innovatiestudies met betrekking tot de meest controversiële kenmerken
van Mode 2 kennisproductie: transdisciplinariteit, reflexiviteit en kwaliteitszorg. deze agenda heeft als
vertrekpunt gediend bij het formuleren van de onderzoeksvraag van dit proefschrift, zoals hierboven
weergegeven.
na deze literatuurstudie is mijn volgende stap een nadere conceptualisering van (de strijd om) rele-
vantie in termen van mijn theoretisch kader. hoofdstuk 3 ontwikkelt de eerste twee bouwstenen van dit
kader: de geloofwaardigheidcyclus en het contract tussen wetenschap en samenleving. op basis van een
kort overzicht van het gebruik van de contractmetafoor in de literatuur, stel ik voor om deze als een heu-
ristiek te gebruiken om de veranderende verhouding tussen wetenschap en samenleving te bestuderen.
ik stel voor om de inhoud van het contract te beschouwen in termen van drie hoofdcomponenten,
die respectievelijk de identiteit van wetenschap, de beweegredenen om haar te steunen, en de omstan-
digheden waaronder onderzoekers werken behandelen. het tweede element van mijn theoretisch kader
is de geloofwaardigheidcyclus. Middels een korte casestudie van de nederlandse academische chemie
wordt de bruikbaarheid van deze twee concepten geïllustreerd. in deze casestudie verken ik hoe verand-
erende opvattingen van ‘relevantie’ in het contract tussen wetenschap en samenleving invloed uitoefenen
op de wetenschapspraktijk, met speciale aandacht voor de instituties die de geloofwaardigheidcyclus
beheersen. deze analyse geeft een eerste indicatie van de spanning tussen de toenemende druk om veel
te publiceren en de toenemende druk om maatschappelijk relevante resultaten – de laatste jaren steeds
meer gedefinieerd als praktische toepassingen - te genereren.
de volgende drie hoofdstukken (4 t/m 6) gebruiken en verrijken (elementen van) dit kader in de
analyse van casestudies van achtereenvolgens de nederlandse chemie, biologie en landbouwweten-
schappen. elke studie behandelt soortgelijke vragen rondom de veranderende strijd om relevantie.
hoofdstuk 4 presenteert een uitgebreidere studie naar de chemie-casus. tussen 1975 en 2005 kun-
nen vier belangrijke veranderingen worden aangewezen in haar wetenschap-samenleving contract:
1. de financiële middelen voor universitair onderzoek bieden meer ruimte voor gerichte inspanningen
op toepassingsgerichte gebieden;
2. de interacties met de industrie zijn intensiever geworden;
3. universiteiten proberen een actieve rol te spelen in de valorisatie van resultaten van wetenschappelijk
onderzoek, en
4. systematische evaluaties zijn uitgegroeid tot een machtige institutie rondom wetenschappelijk
onderzoek.
de gevolgen van deze ontwikkelingen voor de strijd om relevantie van onderzoekers verschillen tus-
sen de drie bestudeerde chemische vakgebieden. Praktische toepassingen leveren bijvoorbeeld veel
minder geloofwaardigheid op in de milieuchemie, waar toepassingsgerichte onderzoeksagenda’s helpen
om financiering te verwerven, maar niet om prestigieuze artikelen te publiceren of om erkenning te
234
Samenvatting
verdienen van collega’s. in de biochemie helpen praktische toepassingen nauwelijks bij het verwerven
van geloofwaardigheid, want dit gebied is nog steeds sterk gericht op fundamentele vragen. de verschil-
len tussen de gebieden kunnen worden verklaard door de aan- of afwezigheid van machtige ‘upstream-
eindgebruikers’, die zich investeringen kunnen veroorloven in wetenschappelijk onderzoek die zich pas
op de lange termijn terugbetalen.
hoofdstuk 5 behandelt veranderingen in het contract en de invloed daarvan op het daadwerkeli-
jke gedrag van universitaire onderzoekers en de inhoud van hun werk. het richt zich met name op
multidisciplinaire samenwerkingen in de paleo-ecologie en toxicologie. dit hoofdstuk laat zien dat het
maatschappelijk contract van biologie een ontwikkeling heeft doorgemaakt die vergelijkbaar is met die
van het contract van chemie. de identiteit van academische biologie is veranderd van een fundamentele
naar een strategische wetenschap; praktische toepassingen liggen steeds vaker ten grondslag aan de
beweegredenen voor het steunen van biologie; de beschikbare financiering is verschoven en onderzoek-
sevaluaties zijn opgekomen. deze ontwikkelingen blijken zowel prikkels als beperkingen voor multidis-
ciplinariteit te genereren die verschillen tussen de twee onderzochte vakgebieden. de toename van mul-
tidisciplinaire samenwerking in de paleo-ecologie valt te begrijpen als een reactie op druk van buitenaf
om meer ‘relevant’ onderzoek te doen, met name op het gebied van klimaatverandering. in toxicologie
zie ik het tegenovergestelde: hier wordt multidisciplinariteit gebruikt om het fundamentele onderzoek
naar fysiologische mechanismen van toxiciteit te versterken. ik concludeer dat multidisciplinaire samen-
werking verschillende doelen kan dienen, uiteenlopend van het verhogen van het praktische nut van een
fundamenteel vakgebied tot het versterken van de fundamentele basis van een toegepast vakgebied.
de centrale vraag van hoofdstuk 6 is: wat is de invloed van veranderende instituties in het onder-
zoeksysteem op onderzoekspraktijken in nederlandse landbouwwetenschappen? dit hoofdstuk gaat in
het bijzonder over drie dierwetenschappen. deze vakgebieden hebben te maken met een sterk dyna-
mische maatschappelijke context. de nederlandse landbouwwetenschappen staan onder druk om de
inhoud en de organisatie van het onderzoek te veranderen om een bijdrage te kunnen leveren aan een
meer concurrerende en duurzame landbouw. in dit hoofdstuk analyseer ik de veranderingen in de insti-
tuties in het landbouwkundig onderzoekssysteem en de consequenties daarvan voor het dagelijkse werk
van nederlandse dierwetenschappers. in de periode die ik heb bestudeerd, valt een verschuiving op van
het consensus-gedreven, top-down ‘ovo-drieluik’ naar een heterogeen kennissysteem. deze verschuiv-
ing brengt twee belangwekkende veranderingen in de institutionele omgeving van het academische
landbouwonderzoek met zich mee: verschuivingen in de beschikbare financiering, en de opkomst van
systematische onderzoeksevaluaties. als gevolg van de genoemde ontwikkelingen zijn in twee vakge-
bieden de interacties met stakeholders toegenomen, maar om verschillende redenen. in dierfokkerij
& genetica (aBg) hebben onderzoekers veelvuldig contact met fokkerijbedrijven voor het uitkiezen
van uitdagende onderzoeksthema’s, het verzamelen van data en de werving van onderzoeksfinanciering.
dankzij deze interacties slaagt aBg erin om voldoende externe financiering te verkrijgen zonder haar
235
Samenvatting
fundamentele onderzoeksagenda in gevaar te brengen. in dierlijke Productiesystemen (aPS) dragen
contacten met boeren alleen bij aan de dataverzameling, terwijl contacten met toegepaste onderzoekers
en met non-gouvernementele organisaties helpen om toegang te krijgen tot financiering. dit lijkt een
kwetsbare combinatie, want aPS heeft moeite met het opbouwen van een coherente onderzoeksagenda
en met het publiceren van voldoende wetenschappelijke artikelen. in het derde bestudeerde vakgebied,
celbiologie, zijn interacties met ‘gebruikers’ juist eerder afgenomen dan toegenomen, want daar is een
neiging om op te schuiven naar meer fundamentele onderzoeksthema’s die een grotere kans bieden op
prestigieuze publicaties, maar die niet van onmiddellijk belang zijn voor de landbouwsector.
in hoofdstuk 7 bied ik een nadere reflectie op het materiaal dat is gepresenteerd in de voorafgaande
hoofdstukken. ik beantwoord de algemene onderzoeksvraag aan de hand van drie sub-vragen.
1. Welke veranderingen kunnen worden waargenomen in de strijd om relevantie van Nederlandse acade-
mische onderzoekers in de chemie, biologie en landbouwwetenschappen, in de periode 1975-2005?
Wie alle bestudeerde vakgebieden samen in beschouwing neemt, kan drie gemeenschappelijke ont-
wikkelingen waarnemen. ten eerste is in de meeste vakgebieden de strijd om relevantie tijdens het
daadwerkelijke onderzoeksproces geïntensiveerd. interacties met mogelijke gebruikers van onderzoeks-
uitkomsten waren in sommige vakgebieden al langer gangbaar, maar in het algemeen is de frequentie en
het belang van dit soort interacties gegroeid. de tweede gemeenschappelijke trend heeft te maken met
de manier waarop onderzoekers erkenning verwerven: gemiddeld genomen is de waarde van praktische
toepassingen als bron van academische erkenning gedaald. Sinds de zeventiger jaren hebben publicatie-
resultaten sociale en economische impact (verder) verdrongen als bron van academische erkenning. ten
derde is de strijd om relevantie tijdens de werving van financiering geïntensiveerd. in alle vakgebieden
melden de onderzoekers die ik heb geïnterviewd dat de aansluiting van hun werk op kennisbehoeften
van maatschappelijke stakeholders steeds belangrijker is geworden voor het verzamelen van voldoende
financiering.
de veranderingen in de strijd om relevantie varieerden ook tussen de vakgebieden. ten eerste, hoew-
el de rol van stakeholders in het academische onderzoeksproces in het algemeen is gegroeid, varieert
de mate waarin zij betrokken zijn bij het verzamelen van data. ten tweede is er variatie in de (beperkte)
mate waarin praktische relevantie wordt beloond in termen van academische erkenning. ten derde helpt
het beloven van maatschappelijke opbrengsten in alle vakgebieden om financiering te verkrijgen, maar
de mate waarin maatschappelijke stakeholders betrokken zijn bij het bepalen van de daadwerkelijke
onderzoeksagenda verschilt. een vierde dimensie betreft de spanning die in sommige vakgebieden is
ontstaan door de combinatie van deze ontwikkelingen. in sommige vakgebieden is er synergie tussen
wetenschappelijke productiviteit en praktische toepassingen omdat interacties met stakeholders niet
alleen helpen bij het werven van financiering, maar ook behulpzaam zijn bij andere geloofwaardigheido-
236
Samenvatting
mzettingen. in andere velden is er een spanning tussen wetenschappelijke productiviteit en praktische
toepassingen omdat daar het versterken van de industriële, medische of agrarische relevantie van je werk
een verwijdering van de centrale academische debatten impliceert. Men zou dan moeten overstappen op
andere modelsystemen of andere onderzoeksvragen die minder geschikt zijn om over te publiceren in
prestigieuze tijdschriften en die minder snel academische erkenning zullen opleveren.
2. Hoe kunnen de overeenkomsten tussen de veranderingen in verschillende vakgebieden worden ver-
klaard?
de trends die gedeeld worden tussen de vakgebieden kunnen worden begrepen als gevolgen van de
combinatie van drie belangrijke veranderingen in de structurele condities rondom het universitair
onderzoek in nederland. de eerste is de diversificatie van financieringsbronnen. tussen 1975 en 2005
is het aandeel van toepassingsgerichte financiering toegenomen. tegelijkertijd zijn er speciale subsi-
dies ontwikkeld voor getalenteerde onderzoekers, die een beloning vormen voor wetenschappers die
bijzonder productief zijn in termen van wetenschappelijke publicaties. de tweede structurele verande-
ring die de strijd om relevantie in alle wetenschappelijke vakgebieden heeft gevormd is de opkomst van
onderzoeksevaluaties, die worden gedomineerd door bibliometrische kwaliteitsindicatoren. de derde
invloedrijke structurele verandering betreft de heersende opvatting over de missie van universiteiten in
nederland. Sinds 1975 zijn autonomie en onafhankelijkheid geleidelijk vervangen door begrippen als
verantwoording, valorisatie en ondernemerschap als de heersende waarden verbonden met universitair
onderzoek.
3. Hoe kunnen verschillen tussen de veranderingen in de verschillende vakgebieden worden verklaard?
een van de opvallendste uitkomsten van dit proefschrift is dat de veranderingen in de strijd om relevan-
tie sterk verschillen tussen vakgebieden. een mogelijke verklaring voor de verschillen tussen vakgebie-
den kan worden gevonden in de combinatie van de mate van convergentie in hun zoekpatroon, hun
traditionele communicatiecultuur en eigenschappen van hun ‘upstream’ eindgebruikers. ten eerste lijkt
het in divergente velden, met een hoge strategische taakonzekerheid waarschijnlijker dat zich niches
ontwikkelen die passen bij de kennisbehoeften van maatschappelijke stakeholders dan in convergente
velden waarin meer consensus bestaat over de wetenschappelijke agenda en ontsnappen hieraan zwaar-
der wordt gestraft.
ten tweede kunnen we de mate waarin praktische toepassingen als bron van geloofwaardigheid die-
nen, begrijpen als we rekening houden met de traditionele communicatiecultuur van een vakgebied. in
vakgebieden met een ‘landelijke’ communicatiestijl waar een lage ‘people-to-problem’-ratio heerst is er
normaal gesproken een lagere citatiedichtheid. dit impliceert dat bibliometrische kwaliteitsindicatoren
237
Samenvatting
een beperkte validiteit hebben. daardoor is het minder waarschijnlijk dat erkenning puur op publicaties
wordt gebaseerd, en waarschijnlijker dat praktische uitkomsten zoals beleidsadvies, patenten of spin-off
bedrijven ook meetellen. in de meer competitieve ‘stedelijke’ vakgebieden, met een hoge ‘people-to-pro-
blem’-ratio en een hoge citatiedichtheid, zullen bibliometrische indicatoren meer doorslaggevend zijn,
niet alleen in formele evaluaties en managementbeslissingen, maar ook in het toekennen van informele
erkenning.
een mogelijke verklaring voor de variatie in de verhouding tussen praktische toepassingen en
wetenschappelijke productiviteit kan worden gevonden in eigenschappen van andere spelers in het
onderzoekssysteem, met name de eindgebruikers van academisch onderzoek. de vakgebieden die het
meest succesvol bleken in het combineren van stakeholder interacties met academische prestaties waren
velden met vermogende en machtige upstream eindgebruikers met een langetermijnvisie op het nut van
wetenschappelijk onderzoek. door het subsidiëren van fundamentele onderzoeksactiviteiten die goede
kansen bieden op high-impact publicaties, steunen de chemische industrie (in het geval van katalyse) en
dierfokkerijbedrijven onderzoekers gedurende de gehele geloofwaardigheidcyclus.
dit onderzoek heeft twee belangrijke implicaties voor het debat over veranderende wetenschapssy-
stemen. ten eerste maakt het duidelijk dat de toenemende druk om te publiceren de druk om praktische
toepassingen te genereren kan tegenwerken. in veel vakgebieden heb ik een spanning waargenomen tus-
sen onderzoeksactiviteiten die nuttig zijn voor maatschappelijke stakeholders en activiteiten die het bes-
te perspectief bieden op prestigieuze wetenschappelijke publicaties. ten tweede geeft mijn werk aan dat
een nadere differentiatie in wetenschaps(beleid)studies nodig is, omdat de dynamiek van wetenschap
veel sterker per vakgebied verschilt dan de meeste literatuur doet vermoeden. Beide bevindingen stellen
de empirische validiteit van enkele centrale beweringen rondom het begrip Mode 2 kennisproductie op
de proef, met name met betrekking tot transdisciplinariteit, reflexiviteit en vernieuwde kwaliteitszorg.
op basis van mijn bevindingen heb ik twee sets aanbevelingen voor wetenschap- en innovatiebe-
leid geformuleerd. ten aanzien van de financiering van universitair onderzoek waarschuw ik voor een
beperkte (kortetermijn-, economische) opvatting van valorisatie. een te sterke nadruk op kortetermijn-
opbrengsten leidt tot ineffectieve financieringsinstrumenten en tot overschatting van de waarde van
universitaire patenten. Bovendien bedreigt zo’n opvatting de levensvatbaarheid van meer fundamentele
onderzoekslijnen, de stabiliteit van universitaire onderzoeksgroepen en de coherentie van hun activitei-
ten. ik beveel daarom aan om:
• wetenschapsbeleid te baseren op verwachte langetermijnopbrengsten in plaats van kortetermijnop-
brengsten,
• de valorisatie-eisen voor subsidies van het nationale onderzoeksfonds nWo te versoepelen,
• een zorgvuldigere kosten-batenanalyse te maken alvorens universitaire uitvindingen te patenteren, en
• de omvang van onvoorwaardelijke lumpsum subsidie voor universitair onderzoek te herstellen en te
bewaken.
238
Samenvatting
Met betrekking tot onderzoeksevaluaties wijs ik op de perverse neveneffecten van bibliometrische kwali-
teitsindicatoren. Zij beperken de opvatting van academische kwaliteit, bevorderen een monocultuur van
wetenschap en veroorzaken inflatie van de waarde van een wetenschappelijk artikel. om deze problemen
op te lossen, zouden bibliometrische indicatoren moeten worden aangevuld, zowel met kwalitatieve
beoordelingen als met indicatoren van sociale en economische impact. ik beveel daarom aan om:
• een rechtstreekse verbinding van financieringsbesluiten aan bibliometrische indicatoren te allen
tijde te vermijden,
• de criteria voor onderzoeksevaluaties te verbreden en ook indicatoren voor maatschappelijke impact
op te nemen (indien van toepassing), en
• de criteria voor het aanstellen en bevorderen van universitair personeel te verbreden: houd ook
rekening met didactische kwaliteiten en met maatschappelijke impact.
239
dankwoord
dit boekje is niet alleen een product van mijn eigen inzet, maar ook van de interacties met heel veel
mensen om mij heen, in utrecht, amsterdam en ver daarbuiten. ik wil op deze plek graag iedereen
bedanken die mij heeft geholpen door inspirerende discussies, adequate feedback en gezelligheid.
allereerst wil ik mijn drie promotoren bedanken voor de fijne samenwerking! Jullie hielpen me het
overzicht te behouden, mijn blik aan te scherpen en de juiste woorden te vinden om mijn bevindingen in
uit te drukken. harro, ik vind het heel inspirerend hoe jij overal pareltjes weet te ontdekken, niet alleen
in wetenschappelijke teksten of datasets, maar ook in het leven in bredere zin. ruud, jij gaf me de ruimte
om mijn eigen keuzes te maken, maar wees me ook regelmatig op de grote lijn, en herinnerde me eraan
waar het allemaal om begonnen is. John, ondanks de iets grotere afstand heb jij ook een flinke steen
bijgedragen door je scherpzinnige commentaar op mijn plannen en producten.
arie rip, jij verdient in dit dankwoord een ereplaats, want je hebt me veel belangrijke ideeën en
teksten aangereikt. Mijn referentielijst bevat meer dan 15 stukken van jouw hand, maar doet jouw intel-
lectuele bijdrage aan dit proefschrift nog steeds te kort.
verder wil ik alle mensen bedanken die ik aan een interview heb mogen onderwerpen. een aantal
van hen kon ik zelfs verleiden om mijn concept-teksten van kritisch commentaar te voorzien: mijn dank
is groot!
ik ben de collega’s van de innovatiestudies groep dankbaar voor de goeie sfeer, constructieve
feedback en de discussies over innovatie, voetbal, politiek, muziek en hypotheken. het was een luxe om
te mogen profiteren van alle ervaringen van de aios die mij voorgingen! een bijzonder woord van dank
aan Sjoerd, Wouter en roald, voor dagelijkse smalltalk, filosofische sessies en strategisch overleg.
i would like to thank everyone i have had inspiring discussions with about the relevance of science,
but i can impossibly list all their names. a number of colleages have helped me a lot with their light
or heavy criticism on (one of) my texts. i want to thank in particular Mathieu albert, otto auranen,
Marlous Blankesteijn, Bram Bos, Jan de Wit, Jochen gläser, finn hansson, Sjoerd hardeman, gaston
heimeriks, ernst homburg, nicole Klenk, laurens Klerkx, chunglin Kwa, grit laudel, anne loeber,
harro Maat, John Parker, Bart Penders, Julia Quartz, ismael rafols, daan Schuurbiers, clare Shelly-
egan, Peter van den Besselaar, Barend van der Meulen, gab van Winkel and Steve Woolgar. een speciaal
240
woord van dank aan arend Zomer voor het gezellige ‘lotgenotencontact’ en de uitwisseling van ideeën
en bevindingen!
Stefan de Jong en floor van der Wind, jullie wil ik bedanken voor het beschikbaar stellen van data uit
jullie afstudeeronderzoek. en Stefan natuurlijk ook voor je bijdrage aan hoofdstuk 5.
als lid van de onderzoeksschool voor Wetenschap, technologie en Moderne cultuur (WtMc)
heb ik met veel plezier aan een aantal workshops en zomerscholen meegedaan. ik wil Sally, els en Wil-
lem bedanken voor hun grenzeloze inzet als opleidingscoordinatoren. alle deelnemende aios: bedankt
voor de gezellige tijd in ravenstein en voor de boeiende discussies! verder bedank ik WtMc voor haar
bijdrage aan de drukkosten van dit proefschrift.
ik bedank het Kennisnetwerk voor Systeeminnovaties en transities (KSi) voor de financiering van
mijn project en voor de interessante discussies tijdens voortgangsconferenties.
Suzanne, dank voor het ontwerpen van de kaft van dit proefschrift!
tot slot wil ik al mijn vrienden en familie bedanken voor hun belangstelling en support. ook
gesprekken met jullie hielden me scherp, al was het maar omdat ik steeds weer oefende om mijn werk
in een paar zinnen uit te leggen! lieve tecla, Thomas, arend, Minouc en robert: bedankt voor jullie
onvoorwaardelijke steun en interesse! nienke, mijn allerliefste, jij was er altijd voor mij, ook als ik het
even helemaal niet meer wist. Maar bovenal ben ik jou dankbaar voor dingen die nog veel belangrijker
zijn dan dit proefschrift.
dankwoord
241
curriculum vitae
laurens hessels was born in groningen (1980). in 1999 he started his studies at university of amster-
dam. after a research project in microbiology at the Swammerdam institute for life Sciences, he got his
Bachelor’s degree in chemistry in 2004. laurens worked as an intern at the national institute for Public
health and the environment (rivM, Bilthoven) and obtained a Master’s degree in environmental
chemistry in 2005. in the same year, he also completed a philosophical Master thesis on the singularity
of science, based on the constructivist philosophy of isabelle Stengers, and graduated in the Philosophy
of Science (with honours).
from 2006 until 2010 laurens worked on the Phd-project presented in this thesis at the innova-
tion Studies group, utrecht university. This project was part of the Knowledge network on System
innovations and transitions (KSi). laurens completed the education program of the graduate school
for Science, technology and Modern culture (WtMc). he served as a student advisor in WtMc’s
teaching committee in 2007 and 2008. Beside his research work, he supervised master students and
provided tutorials in innovation theory, research methods and innovation policy. early 2008 laurens ini-
tiated the utrecht innovation colloquium, a series of guest lectures on the dynamics and governance of
innovation. laurens presented his work at numerous international conferences and gave invited lectures
at Maastricht university, university of amsterdam and York university (toronto).
Since September 2010, laurens works as a researcher at the rathenau institute (den haag).
242
Scientific publicationshessels, l.K., de Jong, S. & van lente, h. 2010, “Multidisciplinary collaborations in toxicology and
paleo-ecology: equal means to different ends” in Collaboration in the New Life Sciences, eds. J.n.
Parker, n. vermeulen & B. Penders, ashgate, london, 37-62.
hessels, l.K. & van lente, h. 2010, The mixed blessing of Mode 2 knowledge production, Science, Tech-
nology & Innovation Studies, 6 (1): 65-69.
hessels, l.K. & van lente, h. 2010, reward research that benefits society, with kudos or even cash,
Nature, 465: 685.
hessels, l.K., van lente, h. & Smits, r.e.h.M. 2009, in search of relevance: the changing contract
between science and society, Science and Public Policy, 36 (5): 387-401.
hessels, l.K. & van lente, h. 2008, re-thinking new knowledge production: a literature review and a
research agenda, Research Policy, 37: 740-760.
van rijnsoever, f.J., hessels, l.K. & vandeberg, r.l.J. 2008, a resource-based view on the interactions
of university researchers, Research Policy, 37: 1255-1266.
hollander, a., hessels, l., de voogt, P. and van de Meent, d. 2004, implementation of depth-dependent
soil concentrations in multimedia mass balance models. SAR QSAR Environ. Res., 15: 457-468
Popular presshessels, l.K. 2008, denk bij evaluatie ook aan het doel van wetenschap. nrc opinie en debat, 6
december: 9
curriculum vitae